Meta's Auto-RecSys Deploys Autonomous Research Agents for Multi-Day Recommender Experiments
_reachsumit · x · 2026-09-11
Meta presents Auto-RecSys, an autonomous research agent system for long-horizon experimentation on industry-scale recommendation models, tackling two challenges: multi-day training feedback loops and fragile, complex infrastructure.
Key designs:
- Distributed asynchronous execution to run parallel experiments across servers;
- Centralized cross-server memory for persistent, recoverable execution across sessions and failures;
- Cognitive-procedural separation, where natural-language skill files steer LLM reasoning while deterministic scripts enforce operational correctness.
A dual-loop self-evolving architecture lets model-specific playbooks accumulate operational knowledge, cutting human effort per experiment cycle.
More from coding & agent
- Magic Patterns' first engineer: Cursor in Slack, a Patterson review agent, repo clones over worktrees — Y Combinator Blog · 2026-10-28
- Dev builds AI job-search app for his wife, she lands a job, and he open-sources it — CurieuxExplorer · 2026-10-28
- Meshy partners with MCP for Blender: agents drop textured 3D models into your scene — sidahuj · 2026-10-28
- Supercharging LLM classification with logprobs: soft classifiers you can threshold-tune — JnBrymn · 2026-10-28
- Codex daily build series day 23: skipping iCloud plugins for a reset — dotey · 2026-10-28
- Cloudflare Access adds fully customizable pages with HTML + Liquid templates — dinasaur_404 · 2026-10-28