I build AI systems at the intersection of machine learning infrastructure, AI governance, and production reliability. I care deeply about making ML systems trustworthy, auditable, and deployable at scale, not just accurate.
Currently focused on BiasOps — policy-as-code infrastructure for AI governance in regulated industries — and contributing to the open-source AI ecosystem.
BiasOps — Policy-as-Code for AI Governance
Exploring infrastructure-native approaches to ML fairness, compliance, and auditability.
Research into treating AI governance as an engineering discipline — version-controlled, testable, and integrated into ML pipelines rather than bolted on as a dashboard.
- Policy-as-code framework:
biasops scan,biasops validate - Coverage across GDPR, EU AI Act, EEOC, NYC Local Law 144, SR 11-7
- Public policy marketplace on GitHub
- Accepted into SCSP AI+Space program
|
Principal ML Engineer · AI Governance & Anomaly Detection Generalized Anomaly Detection Platform (ADaaS) — 10M+ daily transactions across Finance, Audit, Compliance & Legal. Agentic AI monitoring with LangGraph · GPT-4 · DistilBERT · pgvector · MLflow. $50M+ in documented risk avoidance
|
|
Data Scientist · Industrial IoT Predictive models and large-scale data pipelines for manufacturing operations. |
ML Infrastructure → Anomaly detection at scale (10M+ daily transactions)
AI Governance → Policy-as-code, fairness testing, model auditing
Agentic Systems → LangGraph, multi-agent orchestration, LLM ops
NLP / CV → DistilBERT, TextCNN, healthcare AI, HS code classification
MLOps → MLflow, FastAPI, pgvector, serverless AWS, CI/CD pipelines
- $75M+ in risk avoidance — Built AI governance and anomaly detection systems across Finance, Audit, Compliance, and Legal at Schneider Electric
- IEEE Senior Member — TPC member, IEEE FINE 2026 (Track 6: Internet of AI Agents, Osaka)
- Speaker — MLWeek 2026 (ML Governance), SCSP DC 2025, CVPR 2021
- Published — arXiv paper on serverless MLOps for HS code classification (IEEE ICAD 2026)
- Northeastern University — MS Data Science, Outstanding Research Award, RISE 2021
Active Contributor to pydantic-ai — the GenAI agent framework by the Pydantic team.
I write about AI governance, ML infrastructure, and agentic systems on Medium.
Recent pieces:
- Stop comparing AI agent frameworks — here's what actually matters
- Multi-agentic patterns in production
- Bias mitigation beyond preprocessing
Open to conversations about AI governance, ML infrastructure architecture, or BiasOps design partnerships.
LinkedIn · Medium · biasops.ai