Skip to content
View dattgoswami's full-sized avatar

Block or report dattgoswami

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
dattgoswami/README.md

End-to-end systems for autonomous agents — local inference, sandboxed execution, graph and replay memory, MCP, AP2, evals, and continual-learning loops.

  • General Intelligence
  • Ferrum Cloud The improvement loop that plugs into your coding agents and makes them get better over time — capture, replay, distill, share.
  • Free Weights Test-driven coding gym in your browser — TS/Python/Java/Rust, with an offline AI coach (small language model qwen) in your own browser(WebGPU).
  • Continual Intelligence A twelve-part illustrated essay series on the frontier of continual learning and reasoning (covering 600+ papers).
  • Harness Engineering A twelve-part illustrated field guide to building, evaluating, securing, and operating AI agents covering 300+ papers.
  • Agentic AI & LLM Systems — selected technical writing on coding agents and LLM systems.

Pinned Loading

  1. cl-agent cl-agent Public

    Continual-learning substrate for coding agents(Codex, Pi, Hermes, aider, SWE-agent, OpenHands): episode capture, replay, skill distillation, evals, and SOAR/SFT loop.

    Python

  2. ferrum-stack ferrum-stack Public

    Agentic infrastructure for memory, evals, MCP tools, tracing, orchestration, and LLMOps.

  3. ferrum-agent ferrum-agent Public

    Multi-agent orchestration runtime for the Ferrum platform — planning, execution, review, memory, and human approval in one control plane

    Python

  4. ferrum-gateway ferrum-gateway Public

    A standalone service gateway for HTTP, LLM, MCP, and agent traffic, with routing, auth, limits, streaming, observability, token budgets, and tool policy built in.

    Rust

  5. ferrum-evals ferrum-evals Public

    CI and continual-learning evals for Ferrum agents

    Python

  6. ferrum-memory ferrum-memory Public

    Agent working-memory and prioritized experience replay buffer for the Ferrum platform — Qdrant hybrid retrieval, RL experience tuples, and dream-cycle consolidation for continual learning.

    Python