Agentic researcher and entrepreneur building at the intersection of deep tech AI and venture creation. Based in India.
I design cognitive architectures, orchestrate multi-agent systems, and ship infrastructure for autonomous AI — then write about why it matters. My work spans from research-grade agent harnesses to zero-human company tooling, and I'm equally at home in a whitepaper and a pitch.
I also do brand strategy for deep tech founders who want to sound like humans, not brochures.
Currently building foundery.space — a community-ranked directory for founders, researchers, and builders navigating fellowships, grants, accelerators, and residencies. If you're ambitious and building something real, it's for you.
- Building foundery.space — live, growing, shipping weekly
- Contributing to agentic infrastructure tooling (stealth mode)
- Writing on Medium about cognitive architectures and venture building
- Advising two early-stage AI ventures on agentic systems design
Open to conversations about agentic systems, venture building, and deep tech branding. LinkedIn · Email
| Domain | What I'm Actually Exploring |
|---|---|
| Cognitive Architectures | Building memory systems that persist across long-horizon agentic tasks — episodic, semantic, and procedural layers |
| Multi-Agent Orchestration | Designing agent harnesses and swarm coordination protocols for emergent multi-agent behavior |
| Generative Agents | Extending Stanford's generative agents — simulating emergent social dynamics at scale |
| Inference Optimization | Optimizing inference pipelines: speculative decoding, KV-cache management, and edge deployment |
| Agentic Infrastructure | Building the infra primitives — sandboxes, orchestration, and auth — for truly autonomous agents |
| Traversal & Screening Agents | Deploying traversal agents that navigate dense information landscapes and surface signal |
| Regenerative AI | Engineering self-improving agent loops that evolve through repeated interaction |
| Agent Memory Systems | Architecting agent memory systems with principled retention and graceful forgetting |
| Autonomous Research Agents | Building paper-to-agent pipelines that turn research literature into live knowledge systems |
| AI × Entrepreneurship | Building the infrastructure stack for zero-human companies — from orchestration to output |
|
foundery.space Growing to 500+ founders monthly — the directory for navigating fellowships, grants, and accelerators. Community-ranked, actively shipped.
|
GENERATIVE-AGENTS Extended Stanford's generative agents — simulating emergent social dynamics at scale. Interactive behavior in synthetic environments.
|
|
detour Onboard AI agents autonomously saving satellites from orbital debris. Built at TreeHacks 2026.
|
paperclip Orchestration layer for zero-human companies. Shipped, documented, in active development.
|
|
Occium Content System Agentic content pipeline processing 50+ posts/month with automated metrics tracking for LinkedIn + YouTube.
|
Fable 5 System Prompt A curated system prompt library for AI agents. Optimizing LLM behavior through structured, domain-specific instructions.
|
|
VESPER 22 open PRs. Something is being built here.
|
