Name: Applied AI Conf
Start: 2026-05-28T09:00:00+02:00
End: 2026-05-28T18:00:00+02:00
Location: The Delta Campus

May 28, 2026 · The Delta Campus, Berlin · all times CEST (UTC+2)

08:00–09:00
main
Doors Open · Registration & Coffee
09:00–09:10
main
Opening Remarks
09:10–09:40
mainDanila Shtan·Nebius
Opening Keynote
09:45–10:05
mainStanislas Polu·Dust
Building Dust: The Architecture Behind Deploying and Governing Fleets of AI Agents
sideOsman Ramadan·CodeWords
From Writing Code to Trusting Code: How AI Flipped the Engineering Bottleneck
10:10–10:30
mainClara Matos·Sword Health
Beyond Benchmarks: How Evaluations Ensure Safety at Scale in LLM Applications
sideCarl Brenssell·Slack
Always Be Committing: The Scalable LLM Eval Loop
10:30–10:50
expo
Coffee Break
10:50–11:10
mainLucas Hild·Knowunity
Building an AI learning companion: the architecture behind millions of daily interactions
sideClelia Astra Bertelli·LlamaIndex
The Anatomy of LobsterX, a Document Processing Agent
11:15–11:35
mainNico Bentenrieder·Tacto
Procurement Intelligence: When AI Meets Atoms & Bits
sideAlena Astrakhantseva·dltHub
Agents now build 10x more data pipelines than developers. Now what?
11:40–12:00
mainSteffen Hoellinger·Confluent
Context Engineering and Anomaly Detection for event-driven AI Agents with Apache Flink and Kafka
sideKatia Gil Guzman·OpenAI
From Engineer to Orchestrator: How Codex changes the way engineers work
12:05–12:25
mainNeil Zeghidour·Gradium
Giving a Voice to LLMs: Scaling Real-Time Voice Interaction
sideRachel Nabors·Arize
Stop Paying for Frontier Models
12:30–13:45
Kalle Halle
Lunch Break
13:45–14:05
mainSimon Edwardsson·V7
Building Sandcastles for Agents: Safe Execution at Production Scale
13:45–14:15
sideBen Smith·Stripe
Teaching Agents to Pay: What Devs Need to Know
14:10–14:30
mainBalázs Csomor·Langdock
Cache Money: How Prompt Caching Cut Our LLM Bills in Half
14:20–14:50
sideMarouane Khoukh & Mikhail Rozhkov·Nebius
Batch AI Pipelines: How to Go Fast Without Losing Work or Money
14:35–14:55
mainHenry Thompson·Conduct
Deploying Context Graphs into the Fortune 500: Lessons Building the Context Layer for Large Enterprise
14:55–15:25
sideEmmett Fear·Runpod
Inference Without the Wait: A Live Demo of Instant-On Model Deployment
15:00–15:20
mainOğuz Gültepe·Peec AI
Prompt Learning: Distilling Expensive Reasoning Into Fast Production Prompts
15:25–15:45
expo
Coffee Break
15:45–16:05
mainJakob Emmerling·Legora
Building a Universal Agent for Legal
sideLucia Loher & Patrick Löber·Google DeepMind
From Caching to Batching to Flex — How to optimize AI system for production
16:10–16:30
mainSabba Keynejad·VEED.IO
Reinventing VEED for the agentic era
sideJacek Golebiowski·distil labs
The 100x Inference Tax You Don't Have to Pay
16:35–16:55
mainMasashi Beheim·Parloa
Leading through AI change
sideGiselle van Dongen·Restate
Building the Missing Infrastructure Layer for Agents and Distributed Applications
16:55–17:15
expo
Coffee Break
17:15–17:35
mainŁukasz Sągol·Miro
AI-Empowered Engineering Through Collaborative Tooling
sideLucy Zhang·Modal
Build Your Own Background Agent: The infra that scales it to millions
17:40–18:00
mainBruno Show·Choco
Model Routing in Production: What We Learned the Hard Way
sideSacha Uzan·Linkup
Turning the World into Your Context Window: Rebuilding Web Search to Make AI Reliable in Production
18:05–18:35
mainDes Traynor·Fin
Shipping Fin to Production: What Worked, What Broke, What Changed
18:35–18:40
main
Closing Remarks

08:00–09:00

Doors Open · Registration & Coffee

09:00–09:10

Opening Remarks

09:10–09:40

Opening Keynote

09:45–10:05

Building Dust: The Architecture Behind Deploying and Governing Fleets of AI Agents

From Writing Code to Trusting Code: How AI Flipped the Engineering Bottleneck

10:10–10:30

Beyond Benchmarks: How Evaluations Ensure Safety at Scale in LLM Applications

Always Be Committing: The Scalable LLM Eval Loop

10:30–10:50

Coffee Break

10:50–11:10

Building an AI learning companion: the architecture behind millions of daily interactions

The Anatomy of LobsterX, a Document Processing Agent

11:15–11:35

Procurement Intelligence: When AI Meets Atoms & Bits

Agents now build 10x more data pipelines than developers. Now what?

11:40–12:00

Context Engineering and Anomaly Detection for event-driven AI Agents with Apache Flink and Kafka

From Engineer to Orchestrator: How Codex changes the way engineers work

12:05–12:25

Giving a Voice to LLMs: Scaling Real-Time Voice Interaction

Stop Paying for Frontier Models

12:30–13:45

Lunch Break

13:45–14:05

Building Sandcastles for Agents: Safe Execution at Production Scale

13:45–14:15

Teaching Agents to Pay: What Devs Need to Know

14:10–14:30

Cache Money: How Prompt Caching Cut Our LLM Bills in Half

14:20–14:50

Batch AI Pipelines: How to Go Fast Without Losing Work or Money

14:35–14:55

Deploying Context Graphs into the Fortune 500: Lessons Building the Context Layer for Large Enterprise

14:55–15:25

Inference Without the Wait: A Live Demo of Instant-On Model Deployment

15:00–15:20

Prompt Learning: Distilling Expensive Reasoning Into Fast Production Prompts

15:25–15:45

Coffee Break

15:45–16:05

Building a Universal Agent for Legal

From Caching to Batching to Flex — How to optimize AI system for production

16:10–16:30

Reinventing VEED for the agentic era

The 100x Inference Tax You Don't Have to Pay

16:35–16:55

Leading through AI change

Building the Missing Infrastructure Layer for Agents and Distributed Applications

16:55–17:15

Coffee Break

17:15–17:35

AI-Empowered Engineering Through Collaborative Tooling

Build Your Own Background Agent: The infra that scales it to millions

17:40–18:00

Model Routing in Production: What We Learned the Hard Way

Turning the World into Your Context Window: Rebuilding Web Search to Make AI Reliable in Production

18:05–18:35

Shipping Fin to Production: What Worked, What Broke, What Changed

18:35–18:40

Closing Remarks