Raghav Pillai
Raghav Pillai is an individual software engineer and solo developer/creator building at the intersection of vulnerability research and AI. His current public project, OpenTeam, is open-source software for running persistent teams of AI agents on hardware controlled by the user.
At a Glance
- Individuals and technical users who want self-hosted AI agents
- Small teams and businesses running recurring research, operations and administrative workflows
- Developers and organizations requiring local-model or self-controlled infrastructure
- Privacy- and security-conscious users who want data, files, memory and sign-ins on their own server
AI Tools by Raghav Pillai
(1)OpenTeam
Self Hosted AI Agent Workspace
Discussions
No discussions yet
Be the first to start a discussion about Raghav Pillai
Latest News
Published a comparison of Fable 5.2, Gemini 4 Carbon and GPT-6.1 Sol on six OpenTeam agent jobs.
Published a test of Claude Fable 5.2 for AI agents in a self-hosted OpenTeam bot team.
Explained OpenTeam's context-compaction approach for controlling the cost of long agent tasks.
Compared Claude's personal-finance/Plaid capability with OpenTeam's self-hosted Plaid plugin.
Products & Services
Free GPL-3.0 open-source software that runs a team of persistent AI bots on a user's own computer, home server or cloud VM. It includes a server, desktop client and iPhone beta; each bot gets a Linux desktop and browser, shared workspace, memory, routines, plugins and inter-bot messaging.
A simple command-line Git worktree manager listed among Pillai's pinned public repositories.
A TypeScript project listed among Pillai's pinned public repositories; the public profile identifies it by name but does not provide a fuller product description.
Market Position
OpenTeam is positioned as an open-source, self-hosted alternative to hosted agent products such as Grok Bot, Claude Cowork, Perplexity Teammates and Meta Muse, and as an agent-oriented alternative/complement to workflow tools such as n8n. Its differentiators are a computer and browser per bot, persistent memory and sign-ins, multi-agent teamwork, model-provider flexibility and user-controlled hosting. Its tradeoff versus hosted products is that the user must operate the server and supply model access.
Leadership
Founders
Raghav Pillai
Software engineer and independent builder. His public profile describes prior work at Anduril, Amazon, and the University of Texas at Dallas Applied Systems Lab for autonomous-vehicle research; he previously founded Ninestorm and co-founded/led Speck. He holds a B.S. in Computer Science from the University of Texas at Dallas.
Executive Team
Raghav Pillai
Creator and maintainer of OpenTeam; independent software builder
Previously founder/CEO of Speck, software engineer at Anduril and Amazon, and researcher at the University of Texas at Dallas Applied Systems Lab.
Founding Story
Raghav's public career began with Ninestorm, his first entrepreneurship effort, which he says grew to a five-person team and delivered chatbot-based customer-support automation for startups and Fortune 500 companies. After Speck was acquired in January 2026, his public profiles describe a new effort at the intersection of AI and vulnerability research. OpenTeam emerged from that work as an open-source, self-hosted alternative to hosted AI-agent products: the stated goal is agents that keep working on a computer the user owns, with persistent desktops, memory, shared files and user-controlled data.
Business Model
Revenue Model
OpenTeam is distributed as free GPL-3.0 software with no paid tier and no hosted version. Users pay for their own machine or cloud VM, model provider and connected services; the project is not presented as a commercial subscription business.
Pricing Tiers
No paid tier or hosted version; infrastructure, model-provider and connected-service costs are paid directly by the user.
Target Markets
- Individuals and technical users who want self-hosted AI agents
- Small teams and businesses running recurring research, operations and administrative workflows
- Developers and organizations requiring local-model or self-controlled infrastructure
- Privacy- and security-conscious users who want data, files, memory and sign-ins on their own server
- Research and web browsing on sites without APIs
- Meeting preparation and follow-up using email, calendars, meeting notes and CRM/customer context
- Recurring reports, monitoring and scheduled administrative work
- Browser-based workflows that require persistent sign-ins, forms or human takeover for CAPTCHA, 2FA or payments
- Software engineering, code review and server/security checks
- Finance, travel planning, vendor comparison and other multi-step knowledge work