ViennaJS May Meetup
Details
Bring your JS interest and a good vibe! Everybody is welcome & feel free to share this invitation!
Talks and info:
All details: https://viennajs.org/en/meetup/vienna-js-may-2026
Timetable:
- 18:00: Gathering, food & drinks
- ~18:45 - 19:00: Start of the first talk
Talks:
----------------------------------------------
🗣️ Edouard Maleix
▶️ Context That Compounds: What I Learned Building a Memory Lifecycle for Coding Agents
Most teams manage agent context with static rule files — CLAUDE.md, AGENTS.md, custom YAML checklists, skills, etc. I do too. Then I tracked what actually happened: agents repeated the same mistakes across sessions, rules accumulated without evidence they helped, and review load stayed constant no matter how many rules I added.This talk covers what I learned building and testing a different approach: treating context as a living artifact with a lifecycle — generated from real incidents, curated for gaps, compiled to token budgets, and evaluated before injection. The key insight: context you can't measure is context you can't improve.I'll walk through three concrete problems and the patterns I found to address them:
- Attribution — when an agent opens a PR, you can't tell what it wrote, why, or whether a human reviewed it. Giving agents their own signing keys and git identity changes code review from guessing to auditing.
- Memory that persists — Monday you correct an agent, Tuesday it makes the same mistake. I built a typed diary (episodic, procedural, semantic, reflection entries) that survives across sessions and compiles into token-budget context packs. The difference vs. static rules: rules authored from memory can't compound; entries harvested from incidents can.
- Evaluating context, not just code — SWE-bench tests whether agents can fix bugs. I needed something different: does this context pack actually help the agent avoid known mistakes? I'll show why scenarios harvested from real incidents (20–67% baseline → 95–100% with context) wildly outperform auto-generated ones.The talk is grounded in real workflows, real failures, and real data. The patterns — identity, lifecycle-based memory, context-as-testable-artifact — are applicable regardless of which agent or framework you use.
----------------------------------------------
🗣️ Speaker to be announced
----------------------------------------------
🗣️Want to give a Talk?
▶️ https://viennajs.org/en/participate/give-a-talk
Location:
Dynatrace Austria GmbH
THE ICON VIENNA
Tower 24 - 22nd Floor
Wiedner Gürtel 13, 1100 Vienna
When you arrive in the ICON tower, walk straight to the info point and ask for a visitor’s card for Dynatrace (22nd floor). They will explain the way to Tower 24. Walk through the 1st glass door. At the 2nd glass door, you will need your visitor’s card (left side) and hold it against the card reader to open. Hold your visitor card against the card reader at the turnstiles, walk through it, and look at the screen on your slide (while entering). Don’t move too close to the gate before using your card, as it might block you. The screen will display a letter (A-D) about which elevator you need to take. You can also check your designated elevator by holding the card under one of the elevator screen panels. At the elevator entrance is a small display showing you the floor where it is going -> enter if you see the number 22.
Sponsor:
🙏 Dynatrace
Dynatrace definitely is one of Austria’s big success stories in recent history. Founded in 2005 in Linz, we're now one of the fastest-growing SaaS companies on the planet.
Dynatrace is the market leader in software intelligence, that is: We provide full-stack insights into user sessions, applications' runtimes, and infrastructure.
And guess what? We're looking for JavaScript developers: https://careers.dynatrace.com/jobs/?country=Austria&category=Engineering
🙏 Want to sponsor?
We are looking for more sponsors! If you are interested please contact us via viennajsmeetup@gmail.com or at viennajs.org.
***
Don't forget to RSVP!