📍 NYC / SF | 🤖 Applied AI engineer | 🧠 Getting agents into the hands of normies
I take an idea from concept to a production-ready agent - custom integrations, memory, the right context and approvals. Previously at Agno, building how agents reach people. Always down to chat - hiring, exploring or building something together.
- 🔭 Scout - company knowledge agent that navigates Slack, Drive, wiki and CRM instead of chunk-and-embed
- 📊 Dash - self-learning data agent: plain English in, grounded SQL out
- 🧑💻 Coda - code companion that lives in Slack and works from isolated branches
- 🧠 Pal - personal agent that learns how you work, run after run
- 🚀 AgentOS templates - a production agent platform in your own cloud on Railway, AWS or Docker
- 🧬 Learning Machines - owned Agno's learning layer. Agents that learn, not just remember: who you are, what you're working on and what worked last time, so the 1,000th conversation is better than the first
- About you - your profile, preferences and the things you've mentioned before
- About the work - goals, plans and progress in the current conversation
- About the world - companies, projects and people it runs into, plus lessons that help every user
- About itself - a log of the decisions it made and why
- Learns quietly in the background, when it sees fit, or asks first - kept private to you or shared with your team
- 💬 Agents where people work - Slack, Telegram and WhatsApp interfaces with streaming replies and media
- ✋ Human-in-the-loop - agents pause, people approve, in Slack and web frontends
- 🗜️ Context engineering - context compression, tool-history filtering, token counting, context providers
- 🔐 Access control - per-user isolation, scopes and RBAC down to the tool level
- 🔌 MCP and integrations - MCP, Google Workspace, Excel knowledge bases
- 🏢 Fortune 10 company - onboarding, custom workflows, and a better Excel reader and data access for sharper answers from their internal chatbots
- 📚 Knowledge assistant - multi-agent RAG over handbooks for an asset manager, an accounting firm and a law firm, scaled to large document sets with better reranking and more databases
- 📦 Procurement intelligence - seven research agents that reason through each sourcing decision, write the report from the team's template and email it out on their own
- 🚢 Logistics documents - document intelligence with a mixture of specialist agents - far more accurate reads of shipping paperwork, and tedious internal processes automated
- 🧾 Finance and supplier agents - finance and research agents, plus browser agents that log into old government websites, pull the PDFs and summarize them
- 🟢 Open Dots - an open-source personal agent you text; asks before it acts
- 🪟 Open Glass - a company-wide AI workspace with memory people can trust