Background

ML engineer who ships research.

Research: multi-agent coordination, continual learning
Production: ML systems, cloud infrastructure

Research

Multi-agent coordination

Grounded Commitment Learning: verifiable behavioral contracts for AI coordination. Applies Hart-Moore incomplete contract theory. 36.8% hold-up reduction, 382 statistical tests.

Continual learning

Collaborative Nested Learning: multi-timescale optimization with non-adjacent knowledge bridges. +89% accuracy at high regularization. Pareto-dominant across retention-accuracy tradeoff.

Methods

  • Empirical validation with statistical rigor (effect sizes, CIs, multiple hypothesis correction)
  • Formal proofs where applicable (convergence guarantees, conservation laws)
  • Production-quality implementation (95% test coverage, CI/CD, documentation)

Production Experience

Document understanding pipeline

Multi-provider LLM routing with confidence-based escalation. HIPAA-compliant. Human-in-the-loop for low-confidence outputs, RLHF for continuous improvement.

Edge ML

TensorFlow.js models under 25KB. Voice-first field application with Web Speech API + AWS Polly, GPT-powered NLU.

Cross-platform APIs

Minecraft/Roblox/Fortnite integration. State coordination across systems with incompatible assumptions.

Technical Stack

Languages: Python, TypeScript, SQL

ML: PyTorch, TensorFlow, TensorFlow.js

Infrastructure: GCP, Terraform, Docker, CI/CD

Data: PostgreSQL, BigQuery, vector databases

Deployment: Edge (< 25KB models), cloud, hybrid

Education

M.A. Université de Paris VII — Denis Diderot (French-language program)

Littérature, Langues, et Civilisations des Pays Anglophones

National Merit Scholar (Florida State, Emory)

Research Interests

Pre-representational computation—the operations that exist before and enable representation (whether human language, embeddings, or any other representational layer).

  • Which compositions of projection, attention, and regularization preserve structure vs. destroy it?
  • How do systems learn priors over which operations to apply when?
  • Verifiable behavior grounded in observable actions (scalable oversight)
  • Alignment for systems whose operations are themselves learned