AI/ML Engineer · NLP · MLOps · AI Governance
I build machine-learning systems with an emphasis on reliability, traceability, and practical governance. My work spans Python-based ML engineering, production workflows, and the technical implementation of AI accountability.
- ML engineering: Python, model evaluation, APIs, testing, deployment, and MLOps.
- AI governance engineering: audit trails, evidence collection, policy gates, and reproducible verification.
- Research and writing: reliable ML pipelines, AI governance, and the relationship between model outputs and decisions.
- AIGov Compliance Engine — engineering work on auditable AI/ML workflows and evidence verification.
- CPython contribution —
fnmatch/ PathLike contribution (pull request; see GitHub for current status). - Matplotlib contribution —
FixedLocatorhandling of non-finite values (pull request; see GitHub for current status). - SciPy contribution — DIA sparse-format documentation (pull request; see GitHub for current status).
- neuralforecast contribution — contribution and maintainer discussion (pull request; see GitHub for current status).
These links point to contributions, not a claim that every pull request has been merged.
Six public, testable reference implementations focused on trustworthy AI, world models, and reliable ML systems:
| Project | Technical focus |
|---|---|
| Agent Reliability Bench | Agent tool-trace evaluation, batch JSONL reports, confidence intervals; live demo |
| World Models Lab | State-space identification and multi-step rollout baselines |
| Formal AI Verification | Rust approval gates, temporal monitoring, SMT verification |
| Causal ML Benchmarks | Stratified causal effects, propensity weighting, bootstrap intervals |
| Neurosymbolic Reasoning | Horn-rule inference with derivation provenance |
| ML Systems Engineering | Reference and blocked C++ matrix kernels |
Browse the research project showcase.
These are reproducible engineering baselines; they are not presented as peer-reviewed breakthroughs or production-certified systems.
My current interests include reliable machine-learning pipelines, auditability of AI systems, and technical AI governance. For published work and ongoing projects, see my website.