< Speaker / Facilitator >

Zsuzsanna Tamas

Amazon, CoachHub, DKB Code Factory, Lesara, Runtastic, GlobeAir

Buy Ticket

Connect in LinkedIn

Zsuzsanna Tamás is a Senior AI Product Manager at Amazon with 10+ years of experience building recommender systems, AI platforms, and matching algorithms. She’s led product teams at Amazon, CoachHub, DKB, and Adidas Runtastic, and currently drives personalization initiatives at Amazon Music for over 80 million customers. Known for bridging strategy and execution, Zsuzsanna helps teams cut through complexity and adopt AI in ways that are practical, customer- focused, and impactful.

SOLD OUT

Day Overview

Building Agentic AI Applications

16.09

The Social Hub

< About >

Learn to design AI agents that work and make the product decisions that determine whether they stay reliable as they scale. This hands-on workshop takes you from designing a single agent to reasoning about what happens when your system needs a second one: what breaks, what to monitor, and who's accountable. Through a single evolving product scenario, facilitated discussions, and hands-on exercises, you'll build practical frameworks for the decisions PMs face when moving from AI-augmented features to agentic systems.

Best for

Experienced Product Managers building AI-enabled products and customer experiences
PMs who want to upskill from foundational AI concepts to applied agent design and multi-agent thinking
Product Managers working in environments where discovery, decision-making, and execution increasingly happen through AI assistants and automated workflows
PMs looking to build confidence and practical judgment in agentic product development, from a single reliable agent to coordinated systems

Morning session

The morning establishes what makes agentic AI a different product paradigm, not AI that assists, but AI that acts. You'll build a clear mental model for how agents perceive, reason, remember, and take action, and develop the practical LLM literacy needed to make informed product and architecture decisions. The focus then shifts to the design decisions that make or break an agent's reliability: how to scope capabilities, where to place constraints, when to keep humans in the loop, and how to plan for failure in systems that can confidently do the wrong thing. You'll apply these decisions to a realistic product scenario that evolves throughout the day.

< Key learnings >

Understand the shift from AI as a feature to AI as an actor: what makes a system genuinely agentic, and why it changes the product paradigm
Build practical literacy in agent anatomy: how LLMs, memory, tools, retrieval, and APIs fit together
Apply decision frameworks for scoping agent capabilities, designing constraints, and determining where human-in-the-loop or human-on-the-loop oversight is essential
Design for reliability and failure: guardrails, fallback mechanisms, hallucination management, and what it means to trust an agent with real actions
Establish an evaluation and observability mindset from day one: how to know whether your agent is working, and how to debug it when it isn't
Recognise the moment when a single agent is no longer enough and what questions to ask before adding more

Morning session

< Key learnings >

Understand the shift from AI as a feature to AI as an actor: what makes a system genuinely agentic, and why it changes the product paradigm
Build practical literacy in agent anatomy: how LLMs, memory, tools, retrieval, and APIs fit together
Apply decision frameworks for scoping agent capabilities, designing constraints, and determining where human-in-the-loop or human-on-the-loop oversight is essential
Design for reliability and failure: guardrails, fallback mechanisms, hallucination management, and what it means to trust an agent with real actions
Establish an evaluation and observability mindset from day one: how to know whether your agent is working, and how to debug it when it isn't
Recognise the moment when a single agent is no longer enough and what questions to ask before adding more

Afternoon session

The afternoon picks up where success creates new problems. Your agent is working but scope is growing, reliability is under pressure, and a single agent can't cover everything. When do you extend versus split? And once you have multiple agents, what do you need to stay in control? You'll work through these decisions using the same evolving scenario, and close with a forward look at the Agentic Mesh, a conceptual framework for how multiple agents are registered, discovered, monitored, and governed within an organisation.

< Key learnings >

Apply decision frameworks for when to extend an existing agent versus introduce a second specialized one
Understand the core challenges of multi-agent systems: observability, coordination, accountability, and governance across agent boundaries
Explore collaboration strategies and tradeoffs for how agents hand off work, share context, and stay aligned
Build a forward-looking mental model with the Agentic Mesh: registries, discovery, routing, and interoperability and why the design choices you make today shape how well your systems scale tomorrow

Afternoon session

< Key learnings >

Apply decision frameworks for when to extend an existing agent versus introduce a second specialized one
Understand the core challenges of multi-agent systems: observability, coordination, accountability, and governance across agent boundaries
Explore collaboration strategies and tradeoffs for how agents hand off work, share context, and stay aligned
Build a forward-looking mental model with the Agentic Mesh: registries, discovery, routing, and interoperability and why the design choices you make today shape how well your systems scale tomorrow

Roundtable

Conf Overview

The Business Case for Agents: Where to Invest

17.09

CIC Berlin

< About >

Most teams are choosing agentic use cases based on technical excitement or what's easy to demo — not on a clear-eyed economic assessment. This roundtable explores how product leaders are actually evaluating where to invest: what agents really cost, when they pay off, and when a simpler approach is the smarter bet.

< Key learnings >

How to identify high-value use cases versus ones that look exciting but don't survive contact with the economics
The opportunity cost question: what you're not building when you commit to an agent

< Get Speaker's vibes >

Speaker's welcome

< Tickets >

Join the Product Week

Top Tickets

Online

Discover

Conference

See how top teams are really working, and build the network that carries your career.

SEP

Leaders Lab

HoP to CPO

Build a competitive team and AI strategy with Ravi Mehta (Reforge). Leave with a 90-day plan for your org.

ADD LATER

SEP

Craft & AI Labs

Senior PM to HoP

Five workshops, 20 seats each. Get better at building with AI, running product operations and leading the work.

ADD LATER

SEP

Conference

How top teams implement the product model, distribution and how leadership is changing.

INCLUDED

1 main stage + 2 deep dives + 3 roundtables in parallel. Built for exchange, not for sitting and listening.

Save €100

39 left at this price

€499

€399

Inc. VAT

Get Conference

Zsuzsanna Tamas

Morning session

Morning session

Afternoon session

Afternoon session

Join the Product Week

Join the Product Week

Join the Product Week

Join our newsletter to stay up
to date on features and releases.

Join our newsletter to stay up
to date on features and releases.

Zsuzsanna Tamas

Morning session

Morning session

Afternoon session

Afternoon session

Join the Product Week

Join the Product Week

Join the Product Week

Join our newsletter to stay upto date on features and releases.

Join our newsletter to stay upto date on features and releases.

Join our newsletter to stay up
to date on features and releases.

Join our newsletter to stay up
to date on features and releases.