WeAreDevelopers World Congress 2025
2025-07-09..2025-07-11
Berlin, Germany
[WeAreDevelopers World Congress 2019, WeAreDevelopers World Congress 2022, WeAreDevelopers World Congress 2024, WeAreDevelopers World Congress 2025]
who it's for
ai and machine learning engineers, software engineers, platform engineers, engineering managers, frontend developers, developer relations professionals, security engineers
topics
autonomous agent runtimes; GPT-based support agents; code review agents; repo automation CLI; NeMo Guardrails and evaluation; prompt injection detection; module federation micro frontends; PyGame 2D game development; confidential computing and attestation; privacy-enhancing technologies; migrating legacy COBOL with AI; open source governance and licensing
| title | author | length |
|---|---|---|
| The AI Agent Path to Prod: Building for Reliability | Max Tkacz | 32m |
Transforms a naive customer-support agent into a production-ready workflow using n8n and GPT‑4.1, starting from a scoped use case (free‑trial tickets are ~15%… | ||
| Beyond the IDE: A new era of agent collaboration | Ryan J. Salva | 27m |
A new CLI called Gemini CLI ships two weeks ago and is already driving real repo automation: a live demo fixes issue 3740 by creating a branch, writing a… | ||
| AX is the only Experience that Matters | Ivan Burazin | 25m |
Claims that agents will outnumber humans drive concrete changes: 25% of Y Combinator startups report AI wrote 95% of code and 37% build autonomous agents… | ||
| How AI Models Get Smarter | Ankit Patel | 29m |
Three concrete scaling laws drive current gains: pre-training on unlabeled internet text, post-training (fine-tuning and RLHF) to shape behavior, and test-time… | ||
| Advanced Micro Frontends: Multi-Version and Multi-Framework Scenarios | Manfred Steyer | 25m |
Shows concrete recipes: use Module Federation (Webpack) or ESBuild native federation to load micro frontends at runtime, and wrap SPAs as web components via… | ||
| Level Up with Python: Building a Shooter Game Live with PyGame | Louella Creemers | 27m |
A live coding session builds a working 2D shooter with PyGame (notes on PyGame vs PyGame CE) including a movable player, bullets that move ~6 px/frame, enemies… | ||
| Five things in tech that matter and we have to make work | Christian Heilmann | 17m |
Overview of five practical risks and fixes: the old Yahoo Pipes and YQL era shows how fragile mashups become when APIs get closed (Microsoft shutters Bing… | ||
| The Limits of Prompting: ArchitectingTrustworthy Coding Agents | Nimrod Kor | 24m |
A prototype Baz code review agent connects to GitHub via a webhook aggregator, calls cloudset models for structured outputs, and runs concurrent benchmarks on… | ||
| 20 Years of Domain-Driven Design: What I’ve Learned About DDD | Eberhard Wolff | 29m |
Claims that DDD is a 20-year practice focused on steering quality toward a core domain, not spreading it evenly; uses orange stickies in Event Storming for… | ||
| What Makes Open Source Work: Licensing and Beyond | Peter Zaitsev | 25m |
Claims that a license alone is insufficient: projects need license necessary but not sufficient, clear governance (community-run, foundation or neutral rules)… | ||
| Safeguarding Sensitive Data Access At Scale with Privacy-Enhancing Technologies | Mingshen Sun | 24m |
Combines confidential computing enclaves, a KMS plus an attestation service, and serverless compute to isolate only the small functions that touch sensitive… | ||
| The Strategic Power of DevRel: From Community to Business Impact | Emilio Salvador, Ignacio Riesgo, Angie Jones, Nikola Pavesic | 33m |
Panel argues DevRel is essential for developer-facing products and ties value to business metrics like adoption and retention, not just marketing. Companies… | ||
| Beyond the Hype: Building Trustworthy and Reliable LLM Applications with Guardrails | Alex Soto | 29m |
Practical recipe for LLM safety: Java app uses Ollama with Granite 3 Guardian 2B plus classifiers (BERTa zero‑shot, VGG‑based relevance, Code‑Albert… | ||
| Navigating Application Modernisation - Leveraging Gen-AI | Syed Shaaf | 27m |
Static analysis from Konveyor.io flags 149 incidents and Conveyor AI (tool named "Kai") generates concrete code patches, e.g. converting an EJB JMS message… | ||
| Migrating from COBOL with AI: A Moonshot Demo | Julia Kordick | 08m |
An open source project named disassist uses GitHub Copilot, Azure OpenAI and an agent framework to convert COBOL into Java Quarkus; the demo runs locally from… | ||
| Boost your coding productivity with Github Copilot Agent | Alexander Wachtel, Julia Kordick | 29m |
Github Copilot now runs an agent mode inside Visual Studio Code that chooses tools, edits code, runs terminals and iterates until subtasks finish. Demo uses… | ||
| Your Testing Strategy is broken - lets fix it! | Luise Freese | 28m |
Argues that test quantity is a bad proxy and claims 100% coverage gives false confidence, pushing instead for example mapping and behavior-driven development… | ||
| Fireside Chat | Stephanie Cohen | 24m |
Cloudflare sets new defaults on July 1 to block LLM training crawlers for new domains and offers granular controls to block, allow, or charge per crawler via a… | ||
| Software is the New Fuel, AI the New Horsepower - Pioneering New Paths | Katrin Lehmann, Magnus Östberg | 18m |
MB.OS runs in the new CLA as a software‑defined vehicle with over‑the‑air updates and cloud‑service architecture, shipping a high‑performance ECU and an MBUX… | ||
| The future of Stack Overflow as the most vital source for technologists | Prashanth Chandrasekar | 31m |
Stack Overflow says AI adoption in coding rises from about 70% to 80% while trust in AI outputs falls to 29%, and the company now runs Stack Internal used by… | ||
| Compose the Future: Building Agentic Applications, Made Simple | Mark Cavage, Tushar Jain, Yunong Xiao, Jim Clark | 50m |
Docker demonstrates concrete tooling: Compose gains a native Docker model runner keyword to pull OCI-packaged models, Compose wires models and tools to an MCP… | ||
| Outsmarting the System: What Game Cheaters Can Teach Us About Cyber Security | John Romero | 18m |
Games lose an estimated $29B lost revenue annually and studios spend $13.5B on anti-cheat while a shadow $76 million cheatware operation shows cheats run like… | ||
| MCP Mashups: How AI Agents are Reviving the Programmable Web | Angie Jones | 23m |
Revives the old API mashup idea by showing a working pipeline: Model Context Protocol (MCP) servers act as universal adapters so an agent can call Google Maps… | ||
| Fireside Chat | Warren Spector, John Romero | 33m |
Don't imitate others; original ideas and finish something are singled out as the clearest path into games. Indies and festivals like IndieCade, IGF and GDC… | ||
| Building Systems that Last | Werner Vogels | 35m |
Amazon documents a concrete evolution: a monolith becomes service orientation, then 70% key value, 20% single‑table, 10% relational drives DynamoDB and… | ||
| Agents for the Sake of Happiness | Thomas Dohmke | 39m |
Predicts 90% of code will be written by agents, shows Copilot assigning itself an issue, spinning up a VM, cloning a repo and migrating unit tests to pytest… | ||
| nothing matches | ||