Designing and Post-Training Edge Agentic Models: Slides & Talk
maximelabonne · x · 2026-07-31
The author recorded a 30-minute presentation for Cohere Labs Summer School and shared 40 accompanying slides.
The core content focuses on designing and post-training edge agentic models, specifically exploring techniques like Multi-Domain On-Policy Distillation (MOPD) and agentic reinforcement learning.
Related event: Cohere Summer School Talk on Designing Edge Agent Models(2 posts)→
More from coding & agent
- Dev Open-Sources Zeta: An Ambient Agent Runtime Built on Codex — remilouf · 2026-07-31
- Autonomous Prompt and Harness Optimization Drives AI Agent Eval Scores to 100% — saheedniyi_02 · 2026-07-31
- The Hidden Costs of Vibe Coding: Why SaaS is Dead and Lifetime Deals Win — michalmalewicz · 2026-07-31
- Building Enterprise Agents with Google ADK: A Practical Guide — fhinkel · 2026-07-31
- Running Multiple Claude Code Sessions in Parallel: 2 Days of Work in Hours — ai-juzz · 2026-07-31
- Claude Code Catches UI Bug Mid-Build, But Can We Trust Agents Not to Hide Errors? — No-Respect-1040 · 2026-07-31