Agentic AI • ML Platforms • Clean-Tech Optimization
AI Engineer building agentic systems + ML platforms.
I build agentic AI workflows and scalable ML platforms, from data pipelines to model training, evaluation, and production deployment. Recent work spans grid-aware siting optimization, GCP-based ML pipelines, and industry-sponsored research on reliable tool interfaces for AI agents.
Focus
LLM agents, RAG, evaluation, data engineering, cloud deployment
Stack
Python, Spark, GCP, PostgreSQL, Docker, Next.js, Mapbox
Now
Research Assistant (MCP servers) + building grid-aware ML products
Demo clip: Grid-Scout (smart siting engine).
A few work examples
Projects with real systems, data, and constraints
Grid-Scout
Smart siting for gigawatt-scale data centers
A hackathon prototype that helps place large loads where they support the grid: reduce congestion risk, improve reliability, and keep marginal emissions low.
Boeing-Funded MCP Agents for Aero Design Automation
Boeing-funded (NDA), agent orchestration + tool interfaces
Boeing-funded research under NDA focused on making engineering tools to be agent-callable via modular MCP interfaces; public details shared at a high level only.
AI Coach – Multi-Workspace Slack Facilitation Agent
Slack + AWS Lambda + DynamoDB + OpenAI (agentic intervention system)
AI Coach is a serverless, multi-tenant Slack facilitation agent that monitors conversational dynamics and generates lightweight, research-informed nudges using LLMs. Designed for reproducibility and isolation, each deployment runs in its own AWS account, supports multiple Slack workspaces via OAuth, and maintains per-channel rolling state to trigger structured interventions.
FIFA Player Analytics + ML Pipeline
GCP + Spark + PostgreSQL + multi-model regression
A cloud-deployed analytics and modeling pipeline for large, multi-year sports datasets.
Adversarial Robustness Prototype for Autonomous Driving
Hardware demo using Raspberry Pi + Arduino modules
A hands-on prototype that connects adversarial ML insights to a physical autonomy demo.
About
How I build
I like projects where the hard part is the system: messy data, real-time constraints, ambiguous objectives, and shipping something that holds up in production.
- • Build pipelines that are reproducible (tests, schemas, deterministic outputs)
- • Optimize for reliability first, then performance (profiling, caching, batching)
- • Measure quality with evals, not vibes (benchmarks, offline + online metrics)
Beyond work
Things I do when I’m not coding
I like building prototypes, traveling for conferences, and getting out on the water.
