AI Associate Engineer, MLflow

Open-source MLflow, working on how agent runs are recorded and how evaluations are executed.

MLflow, since July 2026

Supporting open-source development of MLflow with a focus on agent observability: trace archival and streaming retrieval, evaluation job execution, and the communication between the server and the client SDK.
Designed a pluggable execution framework for evaluation and scoring jobs, with process isolation, cancellation and timeout semantics held under test.
Partnering with other Red Hat product teams to land their MLflow integrations, translating their requirements into upstream changes and carrying them through public review.
Maintaining Red Hat's downstream MLflow distribution shipped in OpenShift AI across supported release streams: release engineering, CI stability and dependency remediation.

Python / FastAPI / ASGI and async I/O / MLflow / agent tracing / OpenShift AI / Kubernetes / pytest / GitHub Actions

Trusted AI, January to August 2025 (internship)

Built the LLM evaluation metrics surface for TrustyAI: BLEU, ROUGE and Levenshtein.
Rewrote the mathematical foundations of the drift and anomaly detection suite, including a full Python implementation of Jensen-Shannon divergence, and designed the API layer around it.
Architected the service test infrastructure using dependency injection, with validation coverage across every endpoint.

Python / NumPy / SciPy / statistical testing / LLM safety and evaluation metrics / drift detection / dependency injection / Podman / Quay.io