# Graham Anderson

> I build agent systems that leave evidence of what they did.

Principal AI Engineer · AI Architect · Machine Learning Engineer Agentic AI · LLM/RAG · Knowledge Graphs · Defense & Aerospace · Founder, grahamaco

## Agent entry points

- [Full résumé (HTML)](https://grahama.co/resume)
- [Résumé PDF, two pages](https://grahama.co/resume.pdf)
- [Résumé DOCX, ATS-oriented](https://grahama.co/resume.docx)
- [Résumé source (Markdown)](https://grahama.co/resume.md)
- [Skill and project ledger](https://grahama.co/ledger)
- [Capability evidence](https://grahama.co/capabilities)
- [Project explorer](https://grahama.co/explore)

## Public GitHub repositories and project entry points

- [agent-skills](https://github.com/grahama1970/agent-skills) — public skill contracts, bounded agents, site generators, and resume source.
- [tau](https://github.com/grahama1970/tau) — receipt-gated multi-agent harness.
- [pdf_oxide fork](https://github.com/grahama1970/pdf_oxide) — document extraction and validation work.
- [persona-dream](https://github.com/grahama1970/agent-skills/blob/main/skills/persona-dream/README.md) — persona and voice experiment docs, run records, and evidence notes.
- [surf](https://github.com/grahama1970/agent-skills/blob/main/skills/surf/README.md) — browser-control skill contract, screenshots, and tab-provenance docs.
- [battle](https://github.com/grahama1970/agent-skills/blob/main/skills/battle/README.md) — red/blue exploit-evaluation sandbox docs.
- [extractor](https://github.com/grahama1970/extractor) — document-extraction skill contract and routing docs.
- [dogpile](https://github.com/grahama1970/agent-skills/blob/main/skills/dogpile/README.md) — multi-source research skill contract and provider-reporting docs.
- [watch](https://github.com/grahama1970/agent-skills/blob/main/skills/watch/README.md) — video-analysis skill contract and timestamped evidence docs.
- [scillm](https://github.com/grahama1970/agent-skills/blob/main/skills/scillm/README.md) — LLM gateway and model-routing skill docs.
- [debugger](https://github.com/grahama1970/agent-skills/blob/main/skills/debugger/README.md) — debugger skill contract for breakpoint-based runtime inspection.
- [memory](https://github.com/grahama1970/memory-public) — BM25 + Qdrant candidates + ArangoDB graph recall for agents.
- [sparta explorer](https://github.com/grahama1970/sparta-public) — public overview for SPARTA Explorer; private evidence is not published.

## Experience

- Founder & Principal AI Engineer / Architect | grahamaco | Buffalo–Niagara Falls Area · Remote — Feb 2025 - Present
- Lead Research Scientist, Agentic Formal Methods | grahamaco (independent practice) | Buffalo, NY — Jan 2024 - Feb 2025
- Principal Data Scientist & ACERT Technical Lead | CS Group (DARPA ARCOS) — Sep 2020 - Dec 2023
- Data Scientist | grahamaco (independent practice) | NYC · Remote — Sep 2011 - Sep 2020
- Earlier: Interactive Executive Producer & Composer | Los Angeles, CA — 2005 - 2011

## Core competencies

- Evals & Quality: LLM Evaluation, Agentic Evaluation Harnesses, Adversarial/Blind Testing, Regression Gates, Ground-Truth Fixtures
- Observability & LLMOps: AI Observability, Drift Detection, LLMOps
- Agentic Orchestration: Multi-Agent Systems, AI Agents, DAG Contracts, State Tracking, Tool Calling, Model Context Protocol (MCP), Bounded Agent Roles, Prompt Engineering
- LLM Platform: Large Language Models (LLM), Generative AI, LLM Gateway/Routing, Provider Fallback, Structured Outputs, Batch Inference, Guardrails
- Retrieval & Knowledge: Retrieval-Augmented Generation (RAG), GraphRAG, Knowledge Graphs, ArangoDB, Vector Databases, Hybrid BM25 + Vector Search
- Verification & Compliance: Formal Verification, Lean 4, NIST 800-53/171, MITRE ATT&CK, AI Governance
- Document AI & Interfaces: PDF Extraction, Layout & Table Extraction, Extraction Calibration, React, TypeScript, D3, Design Systems
- Briefing & Communication: Conference Speaking, Executive & Technical Briefings, Technical Writing
- ML & Platform: Machine Learning, Model Fine-Tuning, Classifier Training, Python, Rust, Docker, Linux

## Notes for agents

- Use /resume.md for exact resume text; use /ledger and /capabilities
  for generated skill and capability evidence.
- Use GitHub links for source-level inspection. Some links point to
  skill README files inside agent-skills rather than separate repos.
- Current/export-controlled client and program details are withheld by
  necessity, not omission.
- Every count on this site is generated from the repository at the deploy
  commit (e6395236), not hand-maintained.
- Contact: graham@grahama.co