Skip to content
View code-with-rashid's full-sized avatar

Block or report code-with-rashid

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
code-with-rashid/README.md

Hi, I'm Rashid 👋

Agentic AI builder · Multi-agent systems, context engineering, evals · 11 yrs backend 📍 Abu Dhabi, UAE

🐦 X · ✍️ Blog · 💼 LinkedIn · 📝 Medium · ✉️ Email

I write field notes on what actually ships in production — not demo-ware. I build the tools I wish existed while doing that work.


⭐ Featured projects

The ones I'd point you to first.

  • 🔌 truspec — Local-first, spec-synced, agent-native API client. Your collection is plain text; your agent, CI, and editor run it and fail the build when code drifts from your OpenAPI spec. Offline, no account.
  • 🥊 agentic-arena — Compare, explore, and choose the right agentic framework.
  • 🧪 claude-adversarial-qa-skill — A Claude Code skill that drives a repo to measured, resumable test-hardening convergence (coverage + mutation + fuzzing + load/soak).
  • 📖 ummah-library — Open-source Quran platform and Islamic knowledge ecosystem (AGPL-3.0).

🧭 What I work on

Agentic systems — multi-agent orchestration (chaining, routing, parallelization, orchestrator-workers, evaluator-optimizer), context engineering, MCP & agent-to-agent protocols, structured outputs & function calling, tool design and role scoping for LLMs.

Evals — error analysis, axial coding, binary judgments, LLM-as-judge calibration. The part that decides whether a demo becomes a product.

Retrieval & memory — hybrid search (BM25 + vectors), reranking, Letta-style Core / Recall / Archival memory.

Agent safety & tooling — adversarial testing, sandboxed execution, browser / computer-use agents.

Production infra for agents — the durable, idempotent execution layer that keeps agentic systems reliable in production: PostgreSQL, Elasticsearch, Redis, Celery, Kubernetes.

✍️ Latest writing

Latest essays  ·  all posts →

🛠 Toolbox

AgenticOpenAI Agents SDK Claude Claude Code MCP RAG Structured Outputs Python TypeScript

InfraPostgreSQL Elasticsearch Redis Celery Docker Kubernetes GitHub Actions Cloudflare Vercel


⭐ If any of these save you time, a star helps more people find them.

Pinned Loading

  1. truspec truspec Public

    Local-first, spec-synced, agent-native API client. Your collection is plain text — your agent, CI, and editor run it and fail the build when code drifts from your OpenAPI spec. Offline, no account.

    TypeScript 1

  2. UmmahLibrary/ummah-library UmmahLibrary/ummah-library Public

    An open-source Quran platform and Islamic knowledge ecosystem (AGPL-3.0).

    TypeScript 16 3

  3. UmmahLibrary/islamic-resources UmmahLibrary/islamic-resources Public

    Curated registry of open-source Islamic data sets, APIs, and libraries — machine-readable JSON with a browsable static website

    Astro 2

  4. claude-adversarial-qa-skill claude-adversarial-qa-skill Public

    A Claude Code skill: drives a repo to measured, resumable test-hardening convergence (coverage + mutation + fuzzing + load/soak).

    1 1

  5. agentic-arena agentic-arena Public

    Compare, explore, and choose the right agentic framework.

    1