AI agents with a human in the loop

Intelligent systems that stay under human control.

I’m Preetham — I build agentic and on-device AI, and keep a running log of the research frontier from NVIDIA, Anthropic, and beyond.

15
projects
28
tracked papers
on-device
& agentic
Scroll
Agentic AI / On-device LLMs / Gemini / Claude / Nemotron / MCP / Human-in-the-loop / Privacy-first / Reasoning / Playwright / Gemma / Elasticsearch / Agentic AI / On-device LLMs / Gemini / Claude / Nemotron / MCP / Human-in-the-loop / Privacy-first / Reasoning / Playwright / Gemma / Elasticsearch /

Most of my work sits at one of two edges — agents with real autonomy that still ask before they act, and on-device models that keep sensitive data on the device. This site is both the portfolio and the notebook.

01 — Selected work

Things I’ve built

Orca
01 Agentic

Orca

A teamwork graph for AI agents.

Most enterprise AI bolts a chatbot onto a pile of documents and does single-shot retrieval — so it can’t answer the relational questions that actually run a business: “Who do I need to talk to to unblock the Checkout launch?”

Python FastAPI D3 + Cytoscape MCP Azure AI Foundry
View case study
G-Axis
02 Agentic

G-Axis

Your browser already works. G-Axis makes it intelligent.

AI chatbots can answer questions but can’t *do* anything in your browser, and “AI browsers” force you into a separate app — losing your logins, extensions, and habits.

Gemini 2.5 Pro Google ADK FastAPI Playwright Chrome MV3
View case study
03 On-device & Privacy

Weave

Per-viewer product placement, decided on your device.

Ad personalization today either interrupts you or surveils you — per-viewer targeting *requires* shipping your profile to an ad network, the exact data you’d never want to hand over.

React Vite Express Gemma Ollama
View case study
Pravaah
04 Agentic

Pravaah

A multi-agent patient-journey orchestrator.

Hospital care spans dozens of decisions across fragmented systems. Critical patterns get missed when nothing connects the dots — an O₂ saturation slipping 96%→91% over four hours triggers no alarm, but should.

Elasticsearch Kibana Agent Builder ES|QL Elastic Workflows Python
View case study
02 — Research feed

The frontier, tracked

03 — Writing

Notes from the build

Building something at the edge of agents?

I’m always up for a conversation about human-supervised autonomy and on-device AI.

Get in touch