AI Conf 2026 Moscow: Why I'm Attending (and You Should Too)
April 20, Moscow. AI Conf 2026 — a practical Data Science conference with workshops, not just talks.
Why This Conference Matters
400+ participants. Speakers from Sber, Yandex, VK, and other major teams.
The program is unusually applied:
- Multi-agent systems and RAG — how teams actually build and deploy them
- Working with LLMs — achieving stable, predictable results in production
- Justifying tech decisions to business — the eternal challenge
- AI economics — calculating and optimizing infrastructure costs
What I'm Looking For
- Real deployment stories. Not "we used LangChain" — but "here's how we handle failure modes at scale."
- Cost optimization. Every LLM project eventually faces the "why does this cost $50K/month" conversation.
- Agent architectures. Moving beyond simple RAG to systems that can plan, execute, and recover.
The Shift I'm Seeing
Last year, speakers still debated "what's the difference between an LLM application and an agent." Now everyone is building agents. The question changed from "what is it?" to "how do we make it reliable?"
My Plan
Attending at least two workshops:
- Building search agents (hands-on)
- Monitoring with Langfuse (or alternatives)
I'll share notes and key insights on my channel after the event.
More conference notes, production AI insights, and RAG benchmarks from inside a bank — follow my Telegram channel:
🚀 https://t.me/ai_tablet (Russian, technical)
Top comments (0)