City as point cloud
Official city footprints and height fields are simplified at build time, then reconstructed as lightweight returns in the browser.
Project index ready · preparing city scan
SJSU → San Francisco
I build systems
in the Bay.
Six evidence-first products mapped across the places that shaped how I think about infrastructure, research, and reliable AI.
The cities are real. Project placement is a conceptual atlas—not a claim that software has a street address.
- Live demos
- 06
- Replay cost
- $0
- Default
- Keyless
Evidence index
The city has six systems.
Every project has a live, keyless path. Pick a system to see its problem, thesis, architecture, and proof.
The interactive index is loading. Every live project is available directly:
How to read the atlas
Place in one layer.
Proof in the other.
The visual language separates geographic source material from the conceptual relationships between products.
Portfolio as topology
The city recedes. Connections reveal shared primitives: deterministic replay, durable state, evaluation gates, and causal diagnosis.
Two places, one trajectory
SJSU is the origin chapter. San Francisco is where the work continues. The route is autobiographical; project positions are deliberately symbolic.
About Aneesh
Engineering roots in San José.
Building now in San Francisco.
I’m drawn to the infrastructure behind ambitious AI systems: the environments, orchestration, evals, debugging loops, and trust layers that turn a compelling demo into something people can rely on.
More about my work