Now

What I am building

• Hardening the production MLOps pipeline at metricHEALTH: tightening readiness/liveness probes, per-IP rate limits, and the fallback path on the clinical-note summariser.
• Setting up for an incoming MSc in Computer Science at Lakehead University (September 2026): reading, tooling, and shortlisting research directions in scientific ML and interpretability.
• Phase 3 of the cross-survey stellar Gradient Origin Network: the APOGEE DR17, GALAH DR3, and Gaia-ESO DR4 data-engineering layer is done and the model is implemented in PyTorch; the training run at scale is the remaining bottleneck.

Recently shipped

• An essay on FHIR for AI/ML engineers, covering what the spec gets right and where it bites in real pipelines.

What I am reading

• Designing Machine Learning Systems by Chip Huyen, on a slow second pass with the production lens.
• The Qiskit textbook on QSVM kernels, alongside the ICU mortality benchmarking notebooks.

What I am open to

Open to part-time and contract ML work alongside studies, and to research collaboration in scientific ML, interpretability, and ML for healthcare. The fastest path to a conversation is the contact form or a direct email to tirtheshjani@gmail.com.