IbuildAIproductsthatfeelinevitable.
I work as an AI Product Engineer
Software engineer designing and shipping applied-AI products — from retrieval systems and agent runtimes to the interfaces people actually enjoy using.

Engineering with intent

I started writing software because I wanted to understand how things worked. I kept writing it because I found the part I love most: the moment an interface stops feeling like software and starts feeling obvious.
Today I work at the intersection of applied machine learning and product design — building retrieval pipelines, agent tooling, and the front-ends that make them legible to real people.
I care about latency budgets, type safety, motion curves, and the exact weight of a heading. Details are not decoration; they are the product.
The toolkit behind the products
Languages
- TypeScript
- Python
- Go
- Rust
- SQL
AI & ML
- PyTorch
- LangGraph
- RAG systems
- Embeddings
- Evals
Frontend
- React
- TanStack
- Tailwind
- Motion
- Design systems
Backend
- Node
- FastAPI
- Postgres
- Redis
- gRPC
Infrastructure
- Docker
- Kubernetes
- Terraform
- Cloudflare
- CI/CD
Craft
- Motion design
- Accessibility
- Perf budgets
- Observability
Products, not portfolio pieces
Lattice
An interactive embedding explorer that renders million-point vector spaces at 60fps in the browser, with semantic lasso selection and drift monitoring for production models.
- WebGL
- React
- Python
- DuckDB
A record of shipping
Senior AI Product Engineer
Northlight Labs
Lead engineer for the retrieval platform powering every AI surface in the product. Cut p95 answer latency 62% and shipped an eval harness now used by four teams.
- Retrieval
- Evals
- Platform
Full-Stack Engineer
Kestrel
Rebuilt the customer-facing app around a typed end-to-end stack and a shared design system, taking Lighthouse from 54 to 97 and halving front-end defect rate.
- React
- Design systems
- Performance
Software Engineer
Vertex Studio
Shipped data-heavy dashboards and internal tooling for enterprise clients; introduced CI, tracing and load-testing practices that became studio standards.
- Node
- Postgres
- Tooling
Foundations
B.Tech, Computer Science
National Institute of Technology
Focus on distributed systems and machine learning. Graduated with distinction.
Deep Learning Specialization
Independent study
Sequence models, optimization, and production ML systems.
Verified craft
Numbers worth keeping
Let's build something worth remembering.
Open to product engineering roles, applied-AI collaborations, and the occasional ambitious side quest.


