Machine Learning Engineer
Shipping Machine Learning: Prototype → Production
BSCS + MSAI. System design is the center of my work: I architect, build, and operate end-to-end ML and data systems — live pipelines, calibrated models, drift-aware retraining, and the monitoring that keeps them honest in production.
Systems I've designed, built, and run in production.
Smaller builds, across a wide range of domains.
Computer vision, geospatial analysis, probabilistic forecasting, realtime communication, 3D — each of these went from idea to something running.
Gulf Shield AI
Property-risk and insurance analyzer scoring four hazard models into a PDF report.
Poolscope
Detects swimming pools from 0.3m aerial imagery with an NDWI + shape classifier over county parcels.
DroneMap
3D building-height map across eight regions, fusing LiDAR with open building footprints.
Weatherdude
Calibrated probabilistic forecasting bot for daily temperature markets, with its own backtest harness.
Pool Owner Intel
Maps every pool in a metro from assessor records and flags absentee-owned properties.
Pool Routes
Nationwide listing aggregator with multiple-of-monthly-revenue valuation and alerts.
Aegis
Anti-scam payment shield: risky transfers are held until a chosen cosigner approves.
Disqord
Private two-person messenger with WebRTC voice, a shared whiteboard, and voice effects.
Visionboard
Infinite-canvas mind map with a kanban pipeline, pan/zoom, and local persistence.
PadMax
RV/boat storage-facility designer — 2D layout plus 3D walkthrough with a live pro-forma.
Winston
Keyless auction deal-finder that scores surplus and salvage listings locally.
Albatross
Moddable 3D party golf game — twelve courses, physics-driven AI opponents, weather.
I build ML that survives contact with production.
I'm an ML engineer with a BSCS and an in-progress MSAI. What sets my work apart isn't a notebook of experiments — it's systems that run unattended, in production, and correct themselves over time.
Most of what I do day to day is system design: deciding where state lives, how components fail and recover, which guarantees are worth their cost, and what has to be measured for the whole thing to stay trustworthy. Choosing the boundaries well is what makes the rest of the engineering tractable.
I'm most at home where machine learning meets real engineering: streaming data pipelines, feature engineering, model calibration and monitoring, geospatial and computer-vision workloads, and the infrastructure that ties it together. I care about systems that are honest about their own uncertainty and observable when they drift.
- Education
- BSCS · MSAI (in progress)
- Focus
- ML systems · Data · MLOps
- Building
- Production ML, end-to-end
The stack I reach for, grouped by how systems fit together.
System Design
Languages
ML / Modeling
Data & CV
Infra & Full-Stack
Let's build something that runs.
Open to ML engineering, AI, and platform/infra roles. The fastest way to reach me is below — I read everything.
Reach out directly
The contact form activates once a Web3Forms key is added. For now, email is the best path.