Danish Shafi

Product Engineer & AI Builder

I’m a Full-Stack / Product Engineer building across software, backend systems and AI.

I like taking ambiguous problems from idea to working product — designing the architecture, building the interface and shipping the thing. Recently, I’ve been particularly interested in LLM evaluation, agentic systems and the infrastructure behind reliable AI products.

Location
London*
Status
Open to new work

Working on now

Trace

An LLM evaluation harness for analysing model and agent runs, trajectories, grader outputs and failure modes.

Selected Work

  • Workshop

    An automotive operations platform modelling real workshop workflows, with structured service boundaries, relational data and workflow state machines.

    Staff login admin@workshop.compw: password
  • Threads

    An always-on health context platform for chronic illness — connecting patient updates, encounters and evidence-linked observations so every appointment starts with the full story.

  • Winner · Best Use of Cursor
    Replay

    A verifiable multi-agent decision-replay system for urgent care — preserving evidence, recommendations and decision receipts so AI-assisted decisions can be reconstructed and audited.