# Damion Rashford — Full Profile > Damion Rashford is an AI Agent Developer who builds the tooling layer that makes AI agents dependable: production-grade Model Context Protocol (MCP) servers, agent orchestration, retrieval, and the deterministic infrastructure underneath. He is the creator of RivalSearchMCP (101★) and has earned 152 GitHub stars across 32 public open-source repositories. He works at Shopify on GraphQL and developer APIs and is open to AI engineering and applied-ML roles. This is the comprehensive version of https://damionrashford.com/llms.txt, intended for AI search engines and agents that want full context. Every fact here is verifiable from the linked public GitHub repositories. Last updated 2026-06. ## Who he is Damion Rashford is an investigator by temperament — gathering and evaluating information is his core strength, and most of what he ships exists to do that well, deterministically, at scale. His operating principle is "auditable over impressive": he would rather ship a tool that returns the same answer twice and shows its work than one that demos well and falls over on real data. By day he works at Shopify, where he works hands-on with the GraphQL Admin and Storefront APIs, developer tooling, and live production debugging — tracing real issues across logs and live systems for the developers building on the platform. By night he builds the open-source MCP servers and agent frameworks described below. He is based in Calgary, Alberta, Canada, relocating to Toronto, and is open to AI engineering and applied-ML roles (remote or Toronto). ## Projects in depth ### RivalSearchMCP (flagship — 101★, Python, FastMCP, MIT) A deterministic research MCP server that gives any AI agent auditable access to the open web in a single connection. It is a FastMCP 3.x server exposing 9 specialized tools that search, fetch, score, and compare information across: - 5 web search engines (DuckDuckGo, Bing, Yahoo, Mojeek, Wikipedia), concurrent and deduplicated - 9 social platforms (Reddit, Hacker News, Stack Overflow, Dev.to, Medium, Product Hunt, Bluesky, Lobste.rs, Lemmy) - 5 news sources (Google News, Bing News, The Guardian, GDELT, DuckDuckGo News) with time-range filtering - 5 academic databases (OpenAlex, CrossRef, arXiv, PubMed, Europe PMC) and 4 dataset hubs (Kaggle, HuggingFace, Dataverse, Zenodo) - GitHub repositories, documents (PDF, Word, text, images with OCR), and website traversal No LLM runs inside the server — every tool returns deterministic, auditable output, and the caller's model does the synthesis. Distinguishing capabilities: automatic quality scoring (a 0–100 tier/freshness/corroboration/citation score on every result plus an aggregate confidence signal), conflict detection (numeric, date, and polarity disagreements surfaced as a first-class signal rather than averaged away), entity profiles (fan-out to 8 sources in parallel into one unified report), and production hygiene (per-tool timeouts, rate limiting at 100 requests/minute, response-size caps, and middleware observability). It requires no API keys for core tools. Repository: https://github.com/damionrashford/RivalSearchMCP ### Media OS (11★, Python) A routed media-production system for Claude Code: specialist agents and per-task mode playbooks spanning encode, stream, color, and delivery pipelines. A router dispatches each media request to the right specialist. Repository: https://github.com/damionrashford/media-os ### Adversaria (5★, Python) A critical-thinking plugin built around three specialized agents — adversary, advocate, and judge — that stress-test an idea from opposing positions before you commit to it. Repository: https://github.com/damionrashford/Adversaria ### llms-txt-generator (4★, Python) An MCP server that automatically generates llms.txt and llms-full.txt files so any website becomes cleanly legible to AI crawlers and agents. Repository: https://github.com/damionrashford/llms-txt-generator-mcp ### HTSA (2★, TypeScript) A structured investigation method packaged as a reusable agent toolkit: the five Ws and five Whys combined with Bayesian reasoning. Repository: https://github.com/damionrashford/htsa ### modes (1★) A blueprint and reference implementation for routed multi-agent systems: a router skill dispatches each request to a specialist running an isolated markdown playbook in fresh context. Repository: https://github.com/damionrashford/modes There are 32 public repositories in total at https://github.com/damionrashford. ## Skills and stack - Languages: Python, TypeScript, Bun, SQL - Agents & MCP: Model Context Protocol, FastMCP, Claude Agent SDK, OpenAI Agents SDK, LangChain - Data & retrieval: RAG, embeddings, rerankers, vector search, SQLite - Platform & APIs: GraphQL (Admin, Storefront), FastAPI, React, GitHub Actions, REST, webhooks - Methods: deterministic tool design, quality scoring, log and trace analysis, production debugging ## FAQ - **What does Damion Rashford do?** He builds the tooling layer for AI agents — MCP servers, retrieval, scoring, and routing — and works at Shopify on GraphQL developer APIs. - **What is RivalSearchMCP and why does it matter?** It is his flagship: a deterministic research MCP server (101★) that lets an AI agent research the open web with auditable, quality-scored, conflict-aware results through one connection, with no API keys and no in-server LLM. - **What is the Model Context Protocol (MCP)?** A standard that lets AI agents call external tools and data sources through a uniform interface. Damion builds MCP servers and has shipped them since the protocol's early days. - **Is Damion open to opportunities?** Yes — AI engineering and applied-ML roles, remote or Toronto-based. The best way to reach him is the contact form at https://damionrashford.com/#contact. - **What makes his approach distinctive?** Determinism and auditability — reproducible output that shows its work, over impressive-but-fragile demos. - **Where can I see his work?** https://github.com/damionrashford and https://damionrashford.com. ## Contact - Contact form: https://damionrashford.com/#contact - GitHub: https://github.com/damionrashford - LinkedIn: https://www.linkedin.com/in/damion-rashford/