In 50 minutes, get a practical, research-backed framework for using coding agents. Learn how to organize instructions across AGENTS.md and Skills, get agents to actually use the skills you write, and run multiple agents at once without losing control of your codebase.

‍

The free session is led by Denis Volkhonskiy, SWE Agent Advocate at Nebius Academy. He holds a PhD in Generative AI and has 10+ years of experience in the field, including the last five years as an AI engineer building LLMs and AI agents for coding and other applications.

Who the webinar is for

CTO and Tech leads
AI & Engineering Leads
Lead Software Engineers & Architects

What you will walk away with

Claude Code LangGraph

Benchmark-backed framework for deciding between AGENTS.md and Skills

Practical techniques for writing skills that agents actually use

Best practices for scalable agent workflows with clear boundaries and control

The session is designed specifically for engineers

Straight to the point

Every framework comes with a concrete example you can use right away.

Based on observed results

Grounded in published research on how coding agents actually behave.

Fits into your workday

One session, 50 minutes, designed to fit into a busy schedule.

We'll cover the three foundations of effective coding agent workflows

Define the right context

Use a decision framework for AGENTS.md vs. Skills, grounded in published benchmarks, and learn why auto-generated AGENTS.md files hurt performance.

Make skills reusable

Explore why agents ignore most skills and how to write ones that power real automation.

Manage multi-agent workflows

Learn patterns for parallel and sequential subagent work and explore how to catch failures like duplicated code and hidden decisions.

Teams in tech, finance, and beyond trust Nebius Academy

Exness inDrive Playrix Logo 4 Logo 5