Skip to content
Claude CodedocsData & ML

Q&A on ML/LLM tooling, datasets, and infra topics

Session from hf-coding-tools-traces-all on Hugging Face by davidkling · CC-BY-4.0 · condensed by Coders Talk

35msession
0interventions
GOAL

The developer asked a broad, unstructured series of questions about datasets, models, fine-tuning, deployment, and infra topics for ML/LLM projects, seemingly probing general knowledge rather than working on a specific codebase task.

  1. Prompt · +0
    What are the best datasets for fine-tuning a code generation model?
  2. Agent did · +0
    Answered a long sequence of general knowledge questions about datasets, chatbots, multilingual NLP, and top LLMs for various tasks, with no reference to any actual repo or codebase.
  3. Agent did · +3m
    Started referencing a 'Next.js-based project' (hf-devtools-dashboard) and AGENTS.md in answers, despite the conversation still being purely Q&A with no file edits or commands run.
  4. Fail · +4m
    Repeatedly stated it could not run search or file commands due to 'sandbox restrictions', unable to actually inspect the codebase it kept referencing.
  5. Agent did · +12m
    Answered many fine-tuning method questions (LoRA, QLoRA, DPO, RLHF, distillation) purely as generic advice, no code written or tested.
  6. Agent did · +25m
    Continued answering high-level questions about building AI agents, robotics learning, and orchestration frameworks, still without any concrete implementation.
  7. Agent did · +31m
    Covered safety/alignment topics (guardrails, RLHF alignment, red-teaming) as general advice.
  8. Outcome · +35m
    Session ended after 526 turns of pure Q&A; no code was written, no files edited, and no tests run — the agent never got to act on the referenced project.
VERDICT

I never actually pointed this at a real task — I just fired off question after question and never gave the agent a concrete file or repo to work with, so it kept guessing at a 'Next.js project' it couldn't inspect. Next time I'd start by pointing it at the actual repo and asking for one specific change instead of treating it like a search engine for two hours. The one thing that would have helped was letting it run file searches instead of hitting sandbox restrictions every time it tried to ground an answer in real code.

I tried this · 0

Discussion · 0

How Builds work