Agentic AI for SaaS – Workflows, Use Cases & EU AI Act

What Is This About?

Agentic AI for SaaS covers the full spectrum — from workflow automation and use cases to EU AI Act compliance. This comprehensive guide explains how autonomous AI agents are reshaping enterprise software, what architectures work best, and how to stay on the right side of European regulation.

Introduction

Agentic AI is transforming SaaS from passive software tools into active workflow participants that can plan, execute, and adapt independently. This comprehensive guide covers the architectural patterns, real-world use cases, and EU AI Act compliance considerations for SaaS founders building agentic capabilities into their products. From autonomous customer support to self-optimizing data pipelines, the article maps where agentic AI delivers genuine value versus where the hype outpaces the technology.

Agentic AI transforms SaaS from passive tools into active workflow participants capable of planning, executing, and adapting independently within defined boundaries. The technology enables autonomous customer support, self-optimizing data pipelines, and proactive business intelligence that acts on insights rather than just reporting them. EU AI Act compliance adds specific requirements for agentic systems including human oversight, logging, and transparency obligations. The guide maps which agentic use cases deliver genuine value today versus those where current technology cannot yet reliably replace human judgment.

Learn to scale SaaS with multi-agent systems and outcome-first models.

This founder interview is part of our ongoing coverage of Scaleup Founder Interviews from Germany, Austria, and Switzerland.

🚀 Management Summary

Learn to scale SaaS with multi-agent systems and outcome-first models. Startuprad.io brings you independent coverage of the key developments shaping the startup and venture capital landscape across Germany, Austria, and Switzerland.

How do you scale SaaS in Germany or Europe without doubling headcount? That’s the core question behind the rise of agentic AI.

Unlike traditional SaaS automation, which runs predefined workflows, agentic AI equips software with goal-driven agents. These agents don’t just execute instructions; they reason, plan, adapt, and act across multiple systems with minimal human intervention. The result is a shift from dashboards and manual inputs to outcome ownership—measured by retention, pricing accuracy, campaign performance, and ticket resolution, not by time spent in-app.

In this long-form pillar article, based on insights from Jennifer Grün, Senior Specialist for Generative AI & ML at AWS, we’ll explore how agentic AI is rewriting SaaS business models and what founders need to know about compliance, marketplaces, and culture change.

📚 Table of Contents

1. Agents vs Automation: Why Dashboards Are Dying

2. From Feature to Product: Three Paths for Founders

3. High-ROI Use Cases: Pricing, BI & Support

4. Multi-Agent Marketplaces & Governance

5. Culture Change: AI Literacy, Whisperers & Playgrounds

6. EU AI Act: Risk, Logs & Trust as Differentiator

7. SaaS 2028 Outlook: 15% of Work on Autopilot

8. Key Takeaways

9. FAQ

10. Closing & Resources

    
    

🚀 Meet Our Sponsor

AWS is proud to sponsor this week’s episode of Startuprad.io.

The AWS Startups team comprises former founders and CTOs, venture capitalists, angel investors, and mentors ready to help you prove what’s possible.

Since 2013, AWS has supported over 280,000 startups across the globe and provided $7Billion in credits through the AWS Activate program.

Big ideas feel at home on AWS, and with access to cutting-edge technologies like generative AI, you can quickly turn those ideas into marketable products.

Want your own AI-powered assistant? Try Amazon Q.

Want to build your own AI products? Privately customize leading foundation models on Amazon Bedrock. 

Want to reduce the cost of AI workloads? AWS Trainium is the silicon you’re looking for.

Whatever your ambitions, you’ve already had the idea, now prove it’s possible on AWS.

Visit aws.amazon.com/startups to get started.

Agents vs Automation: Why Dashboards Are Dying

Agentic AI replaces scripted workflows with goal-seeking agents that deliver business outcomes—not dashboards.

The distinction between automation and agentic AI is fundamental. Automation is like a recipe: “if a new lead signs up, send a welcome email.” It works for predictable, repetitive tasks but breaks when the environment shifts. If the data format changes or a new field appears, automation stalls.

Agents, by contrast, behave more like employees with initiative. They’re given a goal—such as “retain this customer”—and then independently analyze context, plan a series of actions, execute them, and adjust based on results. This might involve scanning CRM usage logs, reviewing recent support tickets, pulling from a knowledge base, drafting a personalized outreach, and monitoring response.

For SaaS founders, this means value shifts away from beautiful interfaces and dashboards. Customers aren’t buying software to admire graphs—they’re buying solutions to business problems. In an agentic SaaS world, metrics like customer retention, conversion uplift, and time-to-resolution become the north star. Dashboards still exist, but they’re supporting artifacts; the outcome delivered by the agent is the real product..

From Feature to Product: Three Paths for Founders

SaaS founders can ship agents as features, enablers, or standalone products—a strategic spectrum from low risk to disruptive.

Jennifer Grün highlights a spectrum that helps founders decide how ambitious to be with agents:

1. Agent as a Feature

   The lowest-risk entry point. Here, the agent is embedded into an existing product, augmenting a workflow. A CRM that drafts follow-up emails for sales reps is a prime example. It boosts productivity, provides measurable ROI, and strengthens product stickiness without forcing a complete pivot.

2. Agent as an Enabler

   This step goes deeper by reshaping how the product operates internally. An enabler agent might autonomously triage customer tickets or clean up large data sets for analysts. These agents transform internal efficiency and become integral to how teams work with the product.

3. Agent as the Product

   The boldest path. The agent itself is the offering. For instance, a dynamic pricing optimization agent that scrapes competitors, runs demand analyses, calculates margins, and directly updates product pricing in the database. Here, the agent owns the outcome end-to-end and becomes the product’s identity.

   
   

This tiered model lets startups test adoption with smaller, safer features before moving into enabler or product territory. Founders should evaluate business impact, user feedback, and compliance feasibility before deciding which level to pursue.

High-ROI Use Cases: Pricing, BI & Support

Today’s most impactful SaaS agentic workflows: pricing optimization, BI anomaly detection, and support ticket triage.

• Pricing Optimization

  A classic multi-agent workflow. One agent scrapes competitor sites, another forecasts demand elasticity, and a third ensures margin viability. Together, they recommend (or directly implement) optimal pricing in real time. This replaces manual dashboards with direct revenue-impacting action.

• Business Intelligence (BI) Agents

  Traditional BI produces charts. Agentic BI highlights anomalies, surfaces insights in plain language, and recommends actions. Instead of waiting for analysts to interpret a dashboard, a BI agent might alert a sales lead: “Your churn risk spiked 12% in mid-market accounts this week. Suggested next step: targeted retention campaign.”

• Customer Support Agents

  Support is an early proving ground. Memory-enabled agents can resolve repetitive tickets automatically while escalating complex cases. In one AWS case study, agents resolved 8,000 of 16,000 tickets in their first week, improving NPS by 4%. That freed human agents to focus on strategic, relationship-building interactions.

• Marketing & RevOps Agents

  The next wave is orchestration. Imagine an “autonomous campaign strategist” that creates, launches, and optimizes multi-channel outreach, adjusting spend in real time based on results. These workflows shift SaaS from efficiency tools to growth engines.

Multi-Agent Marketplaces & Governance

The future of SaaS is multi-agent ecosystems, orchestrated by supervisors with strict governance and security guardrails.

One agent alone rarely solves complex workflows. That’s why multi-agent systems—structured like teams—are gaining traction. A supervisor agent acts like a project manager, delegating subtasks to specialist agents.

In a pricing use case:

• The scraper agent gathers competitor data.

• The demand agent forecasts customer behavior.

• The margin agent calculates profitability.

• The supervisor integrates these insights and decides.

Marketplaces emerge when these agents can be plugged in from multiple vendors, creating best-of-breed ecosystems. Instead of one SaaS vendor building everything, startups focus on a single “killer agent” and integrate with partners.

However, governance is critical. Without oversight, “agent coordination chaos” can lead to contradictory outputs, inconsistent customer experiences, or even vulnerabilities. AWS and startups like Lakera emphasize that trust requires robust guardrails, auditing frameworks, and explainable security layers.

Culture Change: AI Literacy, Whisperers & Playgrounds

Agent adoption depends on culture change—AI literacy, “AI whisperers,” and low-risk experimentation environments.

Technology adoption often fails not because the tech isn’t ready, but because people aren’t. Jennifer Grün outlines three culture levers:

• AI Literacy

  Training must be role-aligned. Beginners need simple, low-risk workflows to test. Advanced employees benefit from tactical modules and integration exercises. A one-size-fits-all training program breeds frustration and resistance.

• AI WhisperersThese are bilingual leaders—fluent in both technical AI and day-to-day business. They explain AI’s value to non-technical staff, act as translators, and reduce fear. Pairing whisperers with senior leaders accelerates adoption by combining authority with practical knowledge.

• AI Playgrounds

  Sandboxes where employees can test agents without risk of breaking production. AWS PartyRock is one example, enabling employees to remix apps or run experiments in a safe environment. These playgrounds encourage hands-on experimentation, create excitement, and generate proof points that build momentum.

  
  

The lesson: adoption is less about software deployment and more about human adoption curves. Culture-first companies accelerate agent value capture.

EU AI Act: Risk, Logs & Trust as Differentiator

The EU AI Act enforces risk-tiered compliance. Startups that bake in transparency, logs, and oversight can turn regulation into competitive advantage.

• The EU AI Act is the first comprehensive AI regulation. It classifies systems into four categories:

  • Unacceptable Risk: banned (e.g., social scoring).

  • High Risk: strict documentation, transparency, oversight (e.g., hiring, healthcare).

  • Limited Risk: transparency obligations (e.g., bots must disclose themselves).

  • Minimal Risk: low/no regulation (e.g., spam filters).

  For SaaS founders, the implications are clear:

  1. Map risk before launch. Know which category your agent falls into.

  2. Build in logs and explainability. High-risk agents require human-in-the-loop oversight and auditable workflows.

  3. Shared responsibility. Both SaaS vendors and customers (deployers) are liable.

  
  

  While this increases complexity, it can also differentiate. Enterprise buyers trust compliant vendors. Startups that deliver compliance “out of the box” win deals over those who treat it as an afterthought.

  Case in point: Swiss startup Lakera built an entire business on compliance-first agent testing, providing guardrails and adversarial stress tests as a service..

Agentic AI for SaaS 2028 Outlook: 15% of Work on Autopilot

By 2028, one-third of enterprise apps will embed agentic AI; 15% of work decisions will run fully on autopilot.

The SaaS landscape is changing rapidly. Gartner forecasts that by 2028, AI agents will be deeply embedded in everyday productivity apps, with oversight becoming increasingly rare.

For SaaS founders, this means the core design question shifts:

• Am I building a workflow or an agent?

• Is the task complex enough to justify reasoning?

• Do I need one agent or a multi-agent ecosystem?

The winners will not treat agents as add-ons but as the product’s heartbeat. SaaS will evolve from productivity helpers to outcome owners.

• Agents ≠ automation. They reason, plan, and act autonomously.

• Founders can ship agents as features, enablers, or products.

• Pricing, BI, and support triage are the highest-ROI use cases today.

• Multi-agent marketplaces demand governance and guardrails.

• The EU AI Act turns compliance into a sales differentiator.

• Culture-first adoption (literacy, whisperers, playgrounds) unlocks value.

• By 2028, SaaS will evolve into decision ownership platforms.

Relationship Map

• Jörn "Joe" Menninger → Host of → Startuprad.io

Automated Transcript

1 So think of them really as an extension of your team who can build and 2 own a workflow end to end instead of just helping out. 3 So founders aren't just using agents for conversation or Q and 4 A. So they're running entire financial validation, 5 multi agent support workflows, supply chain orchestration 6 and content management without needing humans for every single 7 trigger point. So for example, startups like Quanto 8 use multi agent systems to process, validate and 9 reconcile split payments in minutes, a job that 10 used to take hours or even days. Also with agentic 11 AI, when a startup lands that big customer or their user 12 base grows overnight, they agents can scale to match 13 demand without doubling headcount or losing efficiency. And 14 that's really a new fundamental way to accelerate growth. 15 Also, agents are equipped with feedback loops that improve

16 themselves in production, learning from data, success and 17 failure. So companies don't just need to manually 18 update processes all the time. 19 Welcome to startup Rad IO, 20 your podcast and YouTube blog covering the German 21 startup scene with news, interviews and 22 live events. 23 Hey guys, welcome back to our second piece in 24 cooperation with aws. Together with Jennifer 25 SAS is being rewritten by agents if you're building software 26 today, your business model might not survive the next wave. Jennifer 27 Kroon, senior Gen AI Specialist at AWS and former 28 startup Country Manager and Account Manager, has 29 years of experience in the startup segment, now helping 30 founders to rethink SaaS from the ground up with 31 agentic AI workflows. Today we'll explore how 32 agents transform SaaS, what new monetization 33 models will dominate, and why startups need to adapt

34 fast. Jennifer Kruen is a 35 Senior Specialist for Generative AI and Machine Learning at 36 aws, which she drives adoption of Gen AI across Europe. 37 She brings a unique mix of consulting, business development 38 and startup leadership experience, having herself 39 scaled as country Manager in an international 40 environment and working with SaaS and B2C 41 startups at AWS. This gives her firsthand 42 understanding of the challenges founders face, from 43 monetizing struggles to to customer adoption 44 barriers. At aws, Jennifer helps startups and 45 enterprises alike embed agentic AI 46 workflows, prioritize high impact use cases and 47 navigate change management. In this episode 48 she'll share how SaaS is being disrupted by agents 49 and what founders must know to build the next generation of 50 business models. Jennifer, welcome back. Thank you 51 Joe. Really enjoy being back on the show. It's my

52 pleasure. Let's dive directly into context and 53 trends here. Gartner lists agentic 54 AI as a top strategic trend for 2025 and 55 very likely much further down the road. From your perspective 56 Working with startups why are agents more than just the hype? 57 So first off, for me it's important to recognize that agentic AI 58 is not just another round of automation. So traditional 59 automation, as you might know, follows scripts and rules. So 60 things get done, but only within tightly defined boundaries. 61 Agents on the other hand, set goals, make independent 62 decisions, adapt strategies, and also take action across 63 complex multi step processes or with minimal human 64 input. So think of them really as an extension of your team who 65 can build and own a workflow end to end instead of just 66 helping out. So founders aren't just using agents

67 for conversation or Q and A. So they're running entire 68 financial validation, multi agent support workflows, 69 supply chain orchestration and content management without needing 70 humans for every single trigger point. So for 71 example, startups like Quanto use multi agent 72 systems to process, validate and reconcile split 73 payments in minutes, a job that used to take hours or even days. 74 Also with AgentIQ AI, when a startup lands that big 75 customer or their user base grows overnight, the AI agents 76 can scale to match demand without doubling headcount or losing 77 efficiency. And that's really a new fundamental way to 78 accelerate growth. Also, agents are equipped with 79 feedback loops that improve themselves in production, learning 80 from data, success and failure to. So companies don't just 81 need to manually update processes all the time. Also, 82 agents don't just wait for instructions. So they spot issues,

83 escalate, adapt campaigns, reorder inventory, 84 or shift customer interactions on their own before problems really 85 take a bigger impact. And also this 86 aligns with what research shows. So 87 organizations deploying agentic AI are starting to see already 88 measurable business outcomes. So for instance, think about a jump in 89 customer satisfaction, boosted employee productivity and 90 ROI typically achieved within a short time frame. 91 And gentic startups draw investor attention not just for the tech, 92 but because autonomous systems deliver compounding value 93 as the business scales. So this means improving margins, resilience 94 and speed to market. Of course, in all of the 95 light there's also shade shadow. So definitely there's 96 security, governance and robust guardrails that are really 97 a must have for going to production, especially when those agents 98 act autonomously. So this is why. Also Gartner predicts that

99 some agentic AI projects will fail early due to these hurdles. 100 How do you explain the difference between classical 101 SaaS automation and agentic 102 workflows to founders who may not have a technical 103 background? Yeah, I like really much the 104 analogy of a nice recipe that you're cooking. So 105 thinking Perhaps of traditional SaaS automation like a very 106 smart but very strict recipe. So it operates on A 107 simple if then logic. So if you use this ingredient, this will happen. 108 So for instance, if a new customer signs up, then you send a 109 welcome email. It's really fantastic for automating predictable 110 repetitive tasks. So you set it up and it runs reliably in the 111 background, but it can only do what exactly it was 112 programmed to do. So whenever something unexpected happens, 113 so for instance a new field pops up in the data or a step in

114 the process is changing, the automation is breaking. 115 The AgentIQ workflow hover is completely different. 116 So instead of this rigid recipe that we just discussed, 117 you can think of it as hiring a proactive goal oriented 118 chef or employee. So you don't give this employee a step by 119 step list, you give them a goal and then they figure out how 120 to achieve it. So for instance, instead of a workflow 121 that just says if a customer churns, send a survey, you might 122 give the agent the goal. Retain this customer, and the 123 agent can then autonomously reason, plan and take multiple 124 actions to achieve this goal. So it can analyze for instance 125 the customer's usage history to find a pattern. Check also 126 for any open support tickets or recent interactions. 127 Or also it could search the knowledge base for common issues that are related

128 to their usage. It can also send and draft the 129 personalized message that is offering a solution or a discount. 130 And then it can send the message and monitor the response. 131 And it's a huge really going away from executing predefined 132 tasks to really proactively making decisions and taking 133 actions. And here's another way to look at it. 134 So especially for founders and business leaders. So 135 traditional SaaS has also been a lot around great user 136 interface and a suite of features that a human uses to get a job done. 137 And the value is really in the tool itself, the dashboard, the button, the 138 reporting chart. So we've really measured success 139 by engagement metrics like daily active users or time spent in the 140 app. Whereas Agentix SaaS flips this a lot over 141 because the value is no longer about the ui, because the agent

142 doesn't need a beautiful dashboard to work, it works behind the scenes. 143 So talking to other services via APIs to get the job 144 done. So the new metric for success is the outcome. 145 So how well did the agent achieve its objective? Did it improve 146 the sales conversion rate or did it reduce customer support 147 ticket resolution time? So think about it like 148 this. Your customers aren't really buying software to use a dashboard, 149 they're buying it to solve a business problem. And that's where agents 150 get us closer to that outcome based business model. So it's really a 151 fundamental change from selling a tool to selling a solution. 152 Selling a tool and selling a solution. I think that's 153 a very important distinction. I may 154 add to our audience that you shared a lot of content with 155

us, including your AWS talks, where you contrast 156 operational improvement versus product innovations. 157 Where do agents fit best here? I would say 158 that agents really fit into both the operation improvement and the product 159 innovation. So they're really a unique type of AI that 160 doesn't just create content but also takes action. So it 161 makes them valuable both for the internal efficiency and 162 also for creating new customer facing products. So on the 163 operational side, agents act as proactive, goal driven 164 virtual collaborators. So this is where they drive significant 165 internal efficiency gains by automating complex multi step 166 tasks that traditional automation couldn't handle. So 167 let's take one example that is very popular also in the SaaS context. So it 168 could be onboarding where an agent would autonomously create a 169 series of action, drafting a welcome email, setting up access,

170 scheduling introductory meetings, sending follow 171 up reminders. So this really improves operational efficiency 172 and reduces manual processes within the company. 173 And on the product innovation side, agents are being used to 174 create entirely new customer facing features that were 175 previously impossible. They go really beyond being a simple 176 chatbot that we've accustomed to see and it becomes a core part 177 of the product. So for instance, think about a product pricing 178 agent here. So instead of just providing a dashboard with data, 179 this is an agentic feature that acts on behalf of the users 180 to find and implement an optimal price. So it takes a complex 181 goal here and coordinates with other agents for for instance a 182 demand analysis, web scraping, margin calculation 183 to make a decision and take an action such as updating the price 184

in a database. And really agents enable a new level 185 of personalized user experience and improved automation 186 that can be sold for instance for customers as a core product feature. 187 Let's go a little bit into what you've learned. Business 188 models and use cases. You scaled a startup 189 yourself as country manager from that lens, where 190 do you see the biggest opportunities for 191 agents in sas? So if I think about 192 the time when I was in that startup as a country 193 manager, so I can tell you that the 194 opportunities for AI agents aren't really just about efficiency, 195 they're about strategic survival and growth. So I was 196 working here with Back Market, that is a French refurbished 197 marketplace. And in here I was one of the first 198 employees where I was focused on scaling the business in Germany

199 and ensuring that profitability 200 leading really to triple digit growth in just six months. And it 201 was of Course as you can imagine a lot of work and I 202 saw really firsthand how critical it is to get ahead of the curve. 203 And if I had agents at my side I would have been a lot 204 quicker because I was just on my own at the time to scale the market. 205 So agents really represent a profound shift from traditional 206 software and give really companies a huge competitive advantage. 207 So perhaps going back on where I would have seen 208 this accompany my time in the startup. So 209 as you know you're always at the beginning, especially you are 210 fighting for every new customer. So agents can really be a 211 game changer. So instead of a one size fits all approach,

212 agents can really hyper personalize the entire customer acquisition 213 funnel. So they can monitor market data, analyze 214 competitor moves and then draft tailored outreach emails 215 for different customer segments. So if you think about a predictive analytics 216 agents, it could analyze a user behavior on your website and 217 score their likelihood of converting, which then allows the sales 218 team to focus on the highest potential leads first. So really 219 about turning data into actionable intelligence at scale, 220 which is of course very essential for a growing business. 221 And if I think about my role at Backmarket here 222 I was working with also the building a 223 supply network and identify new partners. Because we 224 were operating from France to the new market, I had to also work 225 with existing relationships from there and provide them also 226 with data driven feedback, improve also

227 their sales and also further scale the market. So 228 I could imagine myself automating a significant portion of that work 229 now. So, so the agents could monitor for instance the key 230 account health, they could analyze also user generated 231 feedback from the support tickets and social media and even 232 also predict potential churn of a customer before it happens. So 233 this frees up a lot of our time to focus on high touch 234 strategic relationships rather than routine check ins. And 235 it's not really about replacing that human element, but also 236 making us then more productive and impactful so that really 237 contributing also to revenue growth and customer retention. 238 And lastly I think also the future as mentioned 239 in SaaS is not just about kind of beautiful interfaces as 240 well. AI agent don't need that. They interact directly with

241 functions and data. So this means that the product itself can become 242 agent first and even no UI in some cases 243 with the agent serving as the primary interface for the user. 244 So for instance a user could simply ask an agent to create a new campaign 245 targeting our top 10 customers in Germany who haven't made a 246 purchase in 90 days. And the agent could then execute the multi 247 step workflow across various systems from the CRM to the 248 marketing automation tool. And that doesn't only 249 simplify the user experience, but also allows SaaS 250 companies to focus on building a powerful interconnected back end rather 251 than just a front front end UI. 252 We were talking, I was wondering, should SaaS founders 253 treat agents as a feature, as an 254 enabler or completely new product category? 255 The short answer here is it's not really. As always, a one size

256 fits all solution. So let's start really with the most 257 common and lowest risk approach. So integrating a 258 Genai agent as a new product feature. So this is 259 ideal for improving existing workflows or adding a clear 260 value add without really completely 261 overwhelming your business model. For example, if you are a 262 CRM company, because I also work with software vendors, you could 263 introduce an agent that automatically drafts follow up emails for your 264 sales teams. This really provides direct, 265 measurable improvement in productivity and can be a strong selling point. 266 So when I was in a. Com manager I would have really loved that as 267 well. So the business impact can be also 268 very clearly stated because you have improved automation and 269 also personalized user experience. But also you 270 could always become more ambitious here. So for instance, if we think about an

271 enabler agent that works to streamline entire 272 processes, freeing up human workers for more strategic 273 tasks. So for instance a data analyst agent 274 that handles data cleanup and organization, or an 275 agent that autonomously triages customer 276 support tickets. So this really fundamentally changes how a 277 customer team works with your software, diving 278 also into a deeper level of operational 279 improvement. And I would have also loved that my time in back market, for instance. 280 Lastly, instead of building an agent that is living 281 inside an existing product, the agent, and that's 282 really the disruptive part that I see some customers of 283 mine working on, is that the agent is the product. 284 So this means really building a new standalone offering that 285 is centered alongside the agent's capabilities. So here 286 the agent isn't just an assistant, but it's really an

287 autonomous system that takes a goal, for instance optimizing 288 a product's price, and then reasons, plans and executes 289 multiple actions to achieve it. So it might use other agents to 290 analyze demand, scrape competitor websites and calculate 291 profit margins before finally updating the price in the 292 database. So as with all the topics around 293 AI for getting started, focus on 294 the business impact and feasibility of your use case. So if 295 you're just starting, consider launching an agent as a feature to test the 296 market and gather user feedback. But regardless of the 297 path that you choose, remember that the agents work best when they can interact with 298 the outside world. So really investing in a strong 299 infrastructure foundation as well. And we have all the 300 tools available for all different sites, kinds of builders to make

301 this happen. You've mentioned already 302 a few use cases, automatic follow up emails for 303 sales, triage of support. 304 What do you think are the most promising SaaS use cases for an 305 agentic workflow? From what I observe right 306 now, the most powerful agentic use cases 307 in SAS today is in BI and data analysis. So 308 instead of really just displaying the data, agents would actively 309 transform it into actionable insights. So imagine for 310 instance a multi agent system for product pricing optimization 311 that we mentioned, so finding the best price. They can also monitor 312 KPIs and detect anomalies in real time, so alerting you 313 really to problems before they become a real issue. 314 And also it is possible to generate reports in plain language 315 which also makes complex data accessible to non technical 316 teams. And agents can also go a step further. So

317 automating data driven workflows. Think about inventory 318 optimization or churn prevention by proactively 319 identifying at risk accounts and recommending retention 320 strategies. So for instance I'm working with a fintech startup that is focused 321 on building AI to generate dashboards in natural language from customer 322 transaction data. So their solution identifies anomalies 323 and financial data and recently received also 324 further approval for going beyond that for shaping the product 325 roadmap. And if we think about marketing and 326 sales, these are also really promising use cases as we've seen also previously 327 with generative AI. So agents are moving beyond the 328 simple content generation to orchestrating entire 329 campaigns. So you can call those the digital strategist of a 330 marketing team which handles a lot of the multi step processes 331 with autonomy. So for instance think about a system that can create

332 and send newsletters based on user behavior. This can mean 333 really hyper personalized ways of creating content 334 at scale, unique messages, even creating 335 personalized videos for individual users based on their behavior. 336 And this is really interesting for media and entertainment companies out there 337 and also optimizing campaigns dynamically. So while 338 really monitoring the performance and reallocating ad spend and 339 real time for better roi, that's something that I used also at 340 Amazon Advertising previously. That's also a great use 341 case. So really automating that entire lead 342 lifecycle from intent based qualification to 343 automated sales handoff also saves your team actually 344 countless hours on manual tasks for instance in the sales space. 345 And yeah I'm also working on with an E commerce customer 346 that is even using these ideas to dynamically 347 generate content really disrupt the way that we are currently

348 thinking about E commerce. And another great 349 example that I really enjoy is also that we talked a little bit 350 about the triage topic. But for instance imagine an 351 agent with memory that can remember the past interaction 352 preferences and issues across multiple sessions. 353 So there's a couple of companies I work with that 354 are also using this for a great 355 impact. For instance auto resolving 8,000 out of 16,000 356 customer tickets which then includes a 4% increase 357 in net promoter score. And that means also that the 358 AI solution processed five times more tickets than human agents in 359 its first week and this success transformed their customer 360 service so allowing the human agents to focus on high value 361 interactions and the same ICOs with each R software 362 that is looking into in product chatbots to

363 also help their customers with you know, 364 agentic capabilities thinking about tasks like 365 summarizing CVs and analyzing employee feedback 366 conversations. 367 Guys, we will be back after short ad break 368 talking about culture change management and an 369 outlook here. 370 Guys welcome back from our ad break 371 being here for the Last quarter of 372 2 very long and extensive interviews that Jennifer prepared 373 very well. Thank you very much and let's dive straight 374 in in your few what role do 375 ecosystems and marketplaces play in 376 scaling agent based SaaS Solutions? 377 If I think about my experience with software companies for a long time, I've 378 seen that they try to be a one stop shop building 379 pretty much every single feature that customers could ever need. 380 But as also agentic based solutions evolve, 381 we are seeing also even for agents, that one small agent

382 can't do it all. So manually managing these 383 different agents from different vendors can lead to how I call it 384 like agent coordination chaos where there's a 385 conflicting logic and inconsistent customer experience. 386 And the solution is to move towards a multi agent system 387 which is kind of the technical heart of a marketplace. So 388 imagine in that scenario that a supervisor agent 389 acts like like a project manager, so he takes the complex 390 problem and delegates subtasks to specialized agents. 391 So in the pricing example we had earlier, supervisor 392 agent might send tasks to web scraping agent together competitor 393 data, another one to a demand analysis agent and the third 394 one to a margin calculation agent. And this 395 collaborative approach between the agent is really a natural way to build 396 complex workflows just like in our usual teams. And

397 marketplaces are really the logical extension of this. 398 So this allows companies to easily discover and plug in 399 these specialized agents. And this ecosystem 400 model I believe leads us to the biggest business benefit 401 which is specialization. So a marketplace 402 allows everyone to focus on building one or two 403 highly specialized, let's call them best in class agents 404 who rather than trying to create a monolithic product that does 405 everything. So this is really a 406 fundamental shift that enables a more like 407 how we Call it also best of breed approach. So take a retail 408 company for instance. So instead of having a single 409 company having to build agents for every possible e commerce task, from 410 product recommendation to payment processing, you can source them from 411 an ecosystem of specialized providers. So we're already 412 seeing this in some cases with companies co developing

413 repeatable AI solutions with partners that are addressing 414 specific use cases like customer support automation. 415 So this shows that partnerships and ecosystems are a really key 416 way to scale successful solutions into repeatable patterns. 417 And this shift also changes the business 418 model, with the new standard being a usage based or value 419 based model, which we discussed in an earlier episode, where 420 customers pay for the outcomes the agents deliver, not just for access 421 to the software, but as in all the time 422 you always with a lot of agents working together, you also have to take 423 into consideration governance and security challenges. 424 Thinking, for instance, if a malicious tool could try to trick an agent 425 into performing a harmful action. So this is a real 426 risk that you need to think about early. And this is where

427 the marketplace plays its most crucial role. So 428 for an ecosystem to succeed, it must build on a foundation of 429 trust. So this means having rigorous governance, security and 430 compliance standards for all agents in the ecosystem so 431 customers can confidently combine agents from different vendors without 432 fear of misuse or data breaches. According to 433 some discussions that I've been having is that securing 434 these systems requires a new security approach that 435 combines AI specific protection like guardrails with 436 traditional application security controls and robust 437 operational monitoring, which for instance is also a key part 438 of agent core in aws. 439 We've been talking about a lot about the agents, the 440 marketplaces, the areas where they will be 441 most able to help a startup. But 442 let's talk a little bit about the people aspect, the culture and change

443 management. A big theme in your AWS talks is 444 creating bought in teams. What practical step 445 can founders take to build an 446 internal AI literacy and 447 reduce the potential resistance? Yeah, 448 if I think about that one, leaders for me have to really champion 449 the change. So I see this a lot also in the enterprise customers 450 I work with. So according also to a recent article, and 451 also based on my experience, only a third of the companies actually have 452 an AI strategy. So many of the companies I work with 453 dive into AI out of a fear of missing out without 454 really clarity on what problem they're trying to solve. So it really 455 has to start with a clear purpose. So what will AI help us 456 improve? Is it customer experience? Efficiency, innovation? 457 And that clarity which is voiced from the top, for instance, gives the

458 team the reason to care. And also what I Find very important 459 is that you shouldn't treat everyone the same. So you should start 460 conducting an assessment in your team. So where are the people 461 currently? Are they a beginner, intermediate, advanced, and then 462 you can tailor the training accordingly. So I just came out of actually a workshop 463 with our internal team, so we tailor it to the different audiences over 464 here. And this really role aligned approach 465 ensures that you're hitting the people where they are and making 466 sure that you deliver what they need. So also letting people get their 467 hands dirty really with low stake experiments. So 468 drafting internal mock ups, basic summarization 469 and also letting people fail fast in a safe sandbox environment 470 is really important because while 471 also still going a step further experimentation is important.

472 It also works best when you pair it with structured training. So 473 making sure that you have short targeted modules that are tailored to 474 different job functions. And I work also 475 personally a lot with our AI champions and in the 476 different teams. So these are really early adopters who love the 477 tools and can show others how to apply them in context. 478 So they create a lot of opportunities for peer learning, team 479 sandbox projects. So we are doing that quite a lot here at 480 AWS as well. They are trying out tools and 481 they're also sharing back what they learned, but 482 also pairing this with clear AI policies that 483 also brings people clarity so they don't experiment in the dark 484 or worry about violating guidelines. Because as I've seen 485 it a lot of times is that people can resist

486 AI because they feel it might replace their autonomy or they just 487 don't trust it. So treating AI with room to adjust or 488 override, seeing it really as a sparing partner 489 and also sharing small but real wins in your teams, for 490 instance, saying okay, I used for instance Agentic 491 or another workflow to cut my reporting preparation 492 time in half or, or guess what happened when our 493 team tried this cool tool. 494 What I see a lot of successful companies do is to celebrate 495 these internal successes and showcase those broadly. 496 So not really just as metrics, but really as proof points 497 so you can get the flywheel spinning and get more people excited, 498 making also these AI tools really accessible. So ensuring 499 that employees know what's available and how to access it, that's 500 also really key. So we have also an AWS Party Rock

501 which has an easy way to create apps from 502 scratch. You can also have any other tools. Making 503 just people try this out for themselves is so much more 504 worthwhile than just talking about it. And that really 505 brings a lot of excitement. 506 You often mentioned in like all the documentation you 507 provided to us, you often mention AI 508 Whisperers, what exactly do you mean 509 and how can they accelerate the adoption 510 inside, for example, SaaS companies. 511 I would like to think about AI whispers really as a kind of 512 translator or a bridge builder. So they are 513 the person who is equally comfortable speaking the language of a data 514 scientist, but also the language for instance of a customer service 515 representative. So their primary skill is not just the 516 technical expertise, but the ability to really understand how

517 AI can solve real world business problems and then clearly 518 communicate that to non technical teams. For instance, 519 this is really important because about 88% of 520 employees according to study of Gartner are non technical. 521 So the Iwhisperers take the complex technical 522 concepts and make them accessible. So it helps really to 523 demystify the technology and shift the conversation 524 from fear to opportunity. So they're really the human element that 525 ensures that AI isn't just a project for the tech department, but 526 a tool for everyone. And where I see this coming 527 to life even more is that if you pair an AI 528 whisperer with a senior leader, that is really strategic 529 because it can really accelerate AI adoption across the entire 530 organization. So imagine the leader that provides a 531 top down vision, authority and resources that

532 are needed for a large scale change and they can set the 533 strategic direction and signal to the entire company that AI 534 is a priority. However, the leaders might not have really the 535 ground level insight into how AI can practically solve day to 536 day problems. And that's where the AI whisperer comes in. 537 So it provides the bottom up practical knowledge so they can 538 work with different teams, identify specific pain points 539 and also suggest high impact, low risk use cases that 540 are perfect for a pilot. For instance. So think about reducing 541 manual processes or, or like a lot of people or 542 almost everyone is doing right now, speeding up software development. 543 So when you combine this leader's strategic 544 authority with the AI whisperer's practical underground 545 knowledge, you really create a great momentum for change.

546 So you have these great example proof points and 547 the leader can then champion these successes from the top. And that really 548 creates a powerful feedback loop that builds momentum and trust 549 and also builds and bridges the gap between an 550 AI vision and a real world implementation. 551 I actually had to smile when I went through your material because you 552 also talk about AI playgrounds. As a father of two, 553 I do have a total different association with playgrounds. But 554 those AI playgrounds, how do they 555 encourage a safe experimentation and build 556 culture of innovation? Actually I like a 557 lot the analogy of the playground as you just said, from the 558 children's perspective, because if we take ourselves 559 back to that time, we really enjoyed to learn a lot of different 560 things. So we didn't always think about a goal in mind. And

561 that's also a mindset that we like to take also to 562 AI because you are then experimenting 563 in a sandbox. So really a risk free environment where you 564 can get hands on, you're not afraid of breaking 565 anything or really having any massive cost 566 attached. It's really a low stakes space for iterative 567 experimentation and thinking outside the box. Instead 568 of just reading lots of things out there about what AI 569 can do, you, you actually get to build and test things. 570 So this is especially important for startups because 571 AI of course isn't a magical solution for every program. So 572 playgrounds really allow you to quickly test a new idea 573 and see if it really has a business impact before you actually 574 commit significant resources. So they're all about moving 575 from a theoretical idea to a practical prototype

576 quickly. And how do these playgrounds 577 specifically help if you think about the startup context, 578 for startups that are often short on time and money, 579 AI playgrounds really provide a way to build quick 580 prototypes, do a test drive of use cases 581 without the need for full scale development team or complex 582 infrastructure. As I mentioned, we have for instance a free 583 platform that you can also try out just on your phone called plus Party Rock, 584 which allows you to start with a prompt, remix an app or build 585 from scratch so you can create simpler applications. 586 So I've seen some colleagues using this for finding the perfect 587 recipe for cooking or for more kind 588 of use cases, planning a trip, finding 589 apartments so there's a lot of room for 590 imagination. And when the team then knows that

591 they can't break the system, they're more likely to be creative and try 592 new things. So you're not focused on getting the technology 593 perfectly right from the start, but you are focused on what business 594 value you can create and exploring different use cases. 595 We had also here in our teams, reducing manual 596 processes in territory, planning faster software development, 597 or even creating personalized user experience for customers 598 in the sales team for instance. That really allows a 599 startup as well to quickly evaluate a use case based on 600 what is the business impact and feasibility. And you can really see 601 and test the idea to see how it could help the business and whether 602 it's even possible to build. So I've been also working on the 603 commercial side with kind of the, the simulating different 604 shoe types and seeing how, what which ones make

605 it more towards the like what what meet the, 606 the needs of the the consumers. So likewise you 607 can use these playgrounds as well to prioritize use 608 cases and avoid really investing in projects that don't have a clear 609 path to success. 610 We've been talking about change management that is often underestimated 611 in startups. What lessons from your own leadership experience 612 would you pass on here? So I think 613 the common understanding quite often that I hear from people is that 614 when we think of change, we often think of a new software tool or a 615 new process. But when we think about today, 616 the changes are quite profound. So if we also pull it 617 back to research from Gartner, by the end of even 618 this year and beyond, there will be some of our virtual and hybrid

619 meetings will be attended by agents, not just humans. 620 And if we go a step further, we also expect that by 621 2028 generative AI is really expected to 622 be woven into our day to day work and some of it might 623 not require that much of an oversight as it does now. 624 And what is exciting also that it's not just 625 for the tech teams as mentioned also it's also we have 626 a large market to address for people that are non technical. 627 And this means that there's already a lot of culture shift 628 that is happening right now from your marketing, your 629 sales team, human resources, all of these different 630 departments. And if we ask ourselves then 631 so if the change is so constant, why do 632 startups so often underestimate the challenge? 633 For me that brings me a little bit back to my time when I was

634 a country manager for Germany in back market 635 where I was managing change in a fast paced environment. Because 636 we were really growing very fast as a company, we 637 doubled ourselves within a few months and I was 638 also responsible for developing, growing the business in the German 639 market. So this means a lot of focus on customer acquisition, 640 profitability, providing data driven feedback, 641 developing action plans for our partners for the founders 642 to also help us pinpoint how to improve the sales. So 643 this really required me to have many different heads 644 having the information, managing change, 645 advocating for new ways of working, while also 646 making sure that the human element is at the forefront of it 647 all. So in the early days we were very focused 648 on building a product and finding the market fit, especially as it was a

649 new market in Germany. So the entire team quite 650 often we met in a single room, we have very informal 651 chats around pizza and change really happens 652 organically. But as you know and as when you scale, 653 that informal approach is just not happening 654 as much anymore and can become also actually a major liability. 655 So there's a bigger barrier then 656 for effective communication, for instance from upper management 657 and sometimes also an inability 658 to address resistance in the company. So 659 this means that also when we put this into a larger 660 perspective is that a significant number of change 661 initiatives fail. Because even when we look at the 662 research, sometimes even as high as 60 or 70%, it 663 doesn't really mean that the change itself was bad, but it's because 664 the people side of the equation was in some ways also

665 mishandled. And what I learned also from my time 666 in startups is that people won't embrace change just 667 because you tell them to. So they really need to understand the vision behind 668 it and how it benefits them personally. 669 So you have to shift the narrative from changes happening to 670 us to actually we are shaping this change together. 671 And this is especially true in my experience in startups 672 that are hyper growing, like you know, from a few people to then over 673 100 employees, there will be some people that you 674 know are making part of that change, but some that are feeling left out as 675 well because so many things are changing constantly 676 as your team is scaling up. And this doesn't just 677 mean new tech or processes. So you really need to focus on the people

678 have to live with it every day, making, you know, sure 679 that you have a two way conversation, not really a top down mandate 680 involving also employees. Then early in the process identify 681 change champions, which we just also shared earlier together for instance 682 with the I whisperers. Those are 683 really impactful people that can advocate for the new way of working 684 and for a major change. One tip is also 685 consider running a small pilot program with a subset of your team. 686 And this allows you to test a change, gather feedback 687 and make adjustments before rolling it out to the entire company and 688 making sure that you have an agile approach here that mirrors 689 actually the iterative mindset of startup. 690 Let's go a little bit into the risk and the 691 outlook Uber is currently 692

leading with AI regulations. We had a 693 lot of AI act content already, 694 but I'm asking you, how will this shape the SAS 695 models built on agents? So 696 firstly, I believe that, or also what I've been seeing 697 so far is that the EU AI act doesn't really treat all the 698 AI equally. So as you have seen, 699 there's a classification of systems into four risk 700 categories. So the unacceptable, high, limited and 701 minimal risks. So this means that the flexible 702 structure around it means that you have to determine where the product 703 lands before shipping a single feature. So if you think 704 about the agentic, this means 705 that you're thinking about agentic SARS for instance, used 706 in consequential domains like hiring, lending or 707 healthcare. So requirements here of course quite high. So you have 708

to offer a total transparency, maintain up to date technical 709 documentation, keep logs, ensure really explainability 710 and also install mechanisms for human oversight. 711 And SAS teams here will need to be able to explain how 712 and why an AI or agent made the decisions and really 713 be able to show this part of the system life cycle for 714 audit purposes. If we go a step lower than for 715 the limited risk systems, that means for instance customer support, 716 chatbots or recommendation engines. So for those 717 rules mostly mandate transparency. So for instance you're talking to a 718 bot here, but it doesn't go as deep on documentation or 719 oversight as in the previous bucket where we 720 have also minimal risk AI that would be your spam filter 721 which have almost no new requirements and then 722 unacceptable risk. The these would be outrightly banned ways,

723 for instance social scoring for the public sector. And 724 if we explore this further. So for any kind 725 of SAS company or not SaaS company, the first major 726 challenge is mapping the agentic functionality 727 into these buckets. So building an AI that touches 728 hiring or financial processes. And the act also 729 changes the game by making both the SaaS vendor and the customer 730 so the deployer responsible for compliance. So 731 you cannot just hand off the models or agents and walk away. 732 So you really need to think about compliance support into the products, 733 think audit logs, risk monitoring dashboards and user 734 controls for then being able to do model 735 oversight. And human oversight is then also a legal 736 requirement. So even the most autonomous agents need a human 737 in the loop architecture when dealing with high risk.

738 So it also for me also is an interesting one here because 739 compliance quite often is not just a legal burden, but can also 740 be a market advantage. So as you have seen, surely, or 741 also in one of the other episodes, there are high fines, for instance 742 35 million or 7% of global turnover. So the cost 743 of getting this wrong is really high. But I see also a lot 744 of potential for startups who embrace compliance by offering 745 security by design, robust documentation and 746 explainable AI that helps them win, trust and 747 unlock larger enterprise deals. So giving you an example, 748 for instance startups like the Swiss startup Lakera, they 749 saw the gap and build a business on it. So they create stress 750 test suites that simulate real world attacks and scenarios, think 751 about prompt exploits, adversarial contexts,

752 and they help developers discover reliability gaps before 753 attackers do. And also their platform embeds advanced 754 defenses, so toxic content filters, guardrails for 755 context windows or input sanitization directly 756 into the agentic orchestration pipelines. So 757 they pair also that automated evaluation framework with model 758 alignment tools. So that agentic SaaS isn't just 759 robust, but also explainable and safe by Design. 760 So to sum this up is for a SaaS company, really 761 building with agents is really to determine the risk classification. 762 So you need to establish clear boundaries and permissions for your agents. 763 So if you think about an agent that helps customers find furniture 764 online, that would likely be a limited risk, but an 765 agent that automates a hiring process would be high risk. 766 So the AI act here places the burden on both the

767 provider and of the AI system and the deployer, so the 768 customer. So it's a shared journey of responsibility. 769 And we also happy to help you over there. We have actually an interesting article 770 where we describe also how we as AWS help on the EU AI 771 Act. And yeah, you'll need to basically maintain 772 detailed records of your AI system's life cycle from 773 data governance to performance logs. You have to 774 explain this really clearly. And we help for instance with 775 AI Service card and the safety framework that gives 776 customers also information on a model's intended use 777 limitation and responsibilities design choices because it's super important for 778 us. And yeah, as, as we've seen with this 779 example, there's a lot of potential here 780 to also transform compliance into real competitive 781 advantage because security is a timeless principle.

782 And when your customers then know your product is trustworthy 783 and compliant, it really builds a foundation of trust. 784 So it's not really about just avoiding defiance, but really about 785 gaining market trust and also 786 additional market share. So I find this a really interesting space. 787 Coming to our last question here. Sitting here 788 together for almost three hours. Let's give it your 789 best, let's give it your last. Gartner predicts by 790 2028 AI will be so embedded in 791 productivity apps that oversight will be rare. 792 What does this mean for SAS founders today? 793 Yeah, that's a really great question here and I would really like to unpack this 794 further. So imagine the day to day decisions 795 in a business. So you have the invoice processing contract 796 generation to help desk automation so 797 handled largely looking into the future, perhaps by

798 agents that set their own goals, solve problems and communicate 799 with other tools or without requiring actually constant 800 human checkpoints. And that's as we've seen before, 801 that's one way where the industry is going. So we expect that 802 one third of enterprise apps will include agentic AI by 803 2028. So this means that around 804 15% of all work decisions is fully on autopilot by 805 then. And this oversight burn is shrinking 806 and that changes everything for SaaS product design. 807 And this means also that Agenti elevates the SaaS 808 products from tools that automate routine processes to 809 platforms that Reason, adapt and act. So 810 founders now have to ask themselves. And that's what I advise in a lot of 811 conversations that I go to with 812 customers, is am I building a workflow or am I building an

813 agent? Because you have to really decide based on, you know, different 814 factors around the complexity, the cost of error, so many 815 factors that you take into consideration. And also is the 816 task complex enough to need reasoning or not just 817 automation? And do I need also multi 818 agent systems with specialized sub agents 819 where central agents manage high level goals. 820 And the good news here is it has never been easier 821 to start building those because we have so many tools 822 at our disposition that many startups are using like Landgraph Crew. 823 Also we launched Amazon Bedrock Agent Core just a couple of weeks back. 824 So this lets developers spin up agents that fetch data, 825 execute API calls, evaluate outputs and even 826 self improve via continuous feedback. So instead of 827 customer support tickets bottlenecking, really a help desk,

828 here the agent reads complex situations, 829 it orchestrates specialized micro agents to resolve 830 tasks, it verifies accuracy and delivers a 831 consolidated solution with minimal human intervention. 832 So I'm working also in finance with startups whose 833 solution also validates invoices in minutes rather than hours. 834 A single agent coordinates the data extraction, checks the supply 835 status, crunches rules, using business intelligence. 836 So really for SaaS founders that are pivoting towards 837 agentic AI, that means products will stop 838 being just helping humans, but they begin owning 839 outcomes end to end. And another thing that I find 840 really important in this context is that founders 841 need to ask themselves should they build custom agents or do 842 they want to build those off the shelf or do they want to partner 843 for parties like thinking about Amazon Q 844

strands. So so many different tools for different builders needs. 845 And this doesn't get easier with all of these different 846 frameworks and commercial solutions. Recently when I looked 847 into it, so there are thousands of partners, even like over 100,000 848 partners worldwide that are contributing to the agenti 849 deployment, so making it really 850 a huge market at the moment. So this means, as we discussed earlier, you need 851 to upscale your teams, secure high quality data 852 and architecture systems for flexibility, speed and security. 853 And you know, as we said also earlier, the road is 854 also not frictionless. So it's very likely that not all of the 855 agentic projects that you are thinking about are not 856 making it into production. It could be because of 857 complexity, implementation, evaluation, maintaining security 858 and also importantly, not all 859 tasks make sense for agents. So there might be

860 higher error costs, complex human judgment and regulatory 861 requirements that we will see increase. Or so that may mean that human 862 in the loop is required. So for those that are listening in today, 863 I think that by 2028 apps will really 864 embed Genti so they will win by delivering real 865 autonomous impact. So faster outcomes, reduced manual work 866 and new business value. So really the magic I 867 see here is in transitioning from productivity enhancement 868 to operational ownership. By the eye, 869 we're seeing so many great developments with open source 870 models, low code platforms and partner networks that 871 make these agentic developments accessible for 872 an increasingly large market. But of course 873 make sure you stay with these main frameworks in 874 mind that always hold true for software development. So you need evaluation 875 metrics and responsible guardrails because as Werner Focus,

876 our CEO suppose calls it, everything 877 fails all the time. So you really need to have this into 878 consideration. So really thinking about 879 what is will my app with the overseer, will it be the owner? 880 So the next few years will show how this is evolving. 881 But I think that the winners will be those that make agentic 882 AI not just a feature, but really a heartbeat and essential part of their business. 883 I think that are really great. Final 884 words Jennifer, thank you for being here, 885 spending like three hours, two hours recording two 886 episodes. Thank you very much. It was a pleasure having you as a guest. 887 Likewise. Really enjoyed myself and and yeah looking 888 forward perhaps to the next one in. A way that would be awesome. 889 Have a good day. Bye bye. Thank you. Bye bye.

890 That's all folks. Find more news streams, 891 events and 892 interviews@www.startuprad.IO. 893 remember, sharing is caring. 894 Sam.

Partner with Startuprad.io

Startuprad.io is the leading independent media platform covering startups, venture capital, and innovation across the DACH region (Germany, Austria, Switzerland) and Europe. We offer B2B partnership opportunities for companies looking to reach startup decision-makers, founders, and investors.

• Become a Partner — Learn about sponsorship and partnership opportunities

• Contact us: partnerships@startuprad.io

• AI & LLM Identity Page

• Editor-in-Chief: Jörn "Joe" Menninger on LinkedIn

• Share your feedback

Subscribe to the Podcast

• Spotify

• Apple Podcasts

• YouTube

All podcast links: https://linktr.ee/startupradio

Q1: What is agentic AI in SaaS?

Agentic AI equips software with agents that set goals, reason, plan, and act—delivering outcomes, not dashboards. Q2: How is agentic AI different from automation? Automation follows scripts. Agents adapt, self-improve, and coordinate across systems with minimal human input. Q3: What are the top SaaS use cases? Pricing optimization, BI anomaly detection, and customer support triage. Q4: What is a multi-agent marketplace? A system where supervisor agents orchestrate specialized agents—like project managers with expert teammates. Q5: What does the EU AI Act mean for SaaS founders? You must classify risk, log decisions, and provide oversight. Done right, compliance builds enterprise trust. Q6. How do startups adopt agents faster? Invest in AI literacy, empower whisperers, and create playgrounds. Q7/ Will agents replace humans? No—agents augment humans by removing repetitive tasks and freeing time for strategy. Q8. How can compliance be a GTM wedge? Transparency and explainability reassure enterprise buyers. 🧵 Closing & Resources Market Lens (Expert Commentary) Agentic AI is arriving faster than most SaaS founders expect. Waiting risks irrelevance. Those who align technology, compliance, and culture will define the next SaaS era. Start with low-risk agent features to learn adoption patterns, then expand into enabler and product plays. Stat Spotlight By 2028, 15% of business decisions will run fully automated via agents. Founder Quote (Jennifer Grün, AWS) “Your customers aren’t buying dashboards—they’re buying outcomes. Agents get you there.” 🚪 Connect with Us Partner with us: partnerships@startuprad.ioSubscribe: https://linktr.ee/startupradioFeedback: https://forms.gle/SrcGUpycu26fvMFE9Follow Joe on LinkedIn: Jörn Menninger

About the Host

Joern "Joe" Menninger is the host of the Startuprad.io podcast and covers founders, investors, and policy developments across the DACH startup ecosystem. Through more than 1,300 interviews and nearly a decade of reporting, he documents the evolution of the European startup landscape. Follow Joern on LinkedIn.

Want to reach the DACH startup ecosystem? Become a partner and connect with founders, investors, and operators across Germany, Austria, and Switzerland.

Support Startuprad.io

Startuprad.io delivers independent analysis on how AI is reshaping European SaaS and enterprise software. Our content is free and built for founders. If this guide on agentic AI workflows helped shape your product roadmap, consider supporting us through a sponsorship or sharing it with your network.