Researcher's agent workflow: label 100 examples, have the agent scale to 10k pseudo-labels
ducha_aiki · x · 2026-10-08
Responding to skepticism that fine-tuning is simpler, duchaaiki describes his months-long workflow: hand-label 100 examples, have an agent find heuristics to extend to 10k pseudo-labels, train models on them, verify predictions manually, and let the agent refine the pseudo-labeling heuristics iteratively.
Related event: Researcher Uses Agents to Scale 100 Labels into 10K Training Samples(2 posts)→
More from coding & agent
- Agents love tidying up files nobody asked them to touch — JFPuget · 2026-10-08
- Open-source bridge turns a browser chat tab into an OpenAI-compatible API endpoint — harshanacz · 2026-10-08
- GraphRAG Bug: Deleted Documents Stay Indexed and Retrievable After Updates — JeremyCMorgan · 2026-10-08
- Paper: Vibe Coding Kills Open Source as AI-Recommended Repos Lose Stars — soumitrashukla9 · 2026-10-08
- Every publishes definitive guide to Compound Engineering, the philosophy behind 7k-star plugin — every · 2026-10-08
- Claude Haiku 5.5 matches GPT-6 Luna pricing but a stingier tokenizer hides a 1.25x cost hike — Simon Willison · 2026-10-08