I'm an independent Staff Applied AI Engineer and product builder. I help teams turn an early idea into an AI product people can use: shaping the experience, building the system, and bringing it to launch.
From leading teams in large organizations to leading agents and delivering impact. My work spans product strategy, hands-on engineering, and the demos and storytelling that help people understand a new product.
CopilotKit is a current client. I'm available for new AI engineering projects, fractional product leadership, and advisory work.
🇺🇸 US · 🇵🇭 PH · Working across borders
Work with me · Portfolio · LinkedIn · Email
Open-source projects I've helped build with CopilotKit, recognized on Trendshift.
| OpenMuse | OpenDots |
|---|---|
| OpenTag | OpenBot |
Live Trendshift badges in the TypeScript category. Each badge links to its source; recognition belongs to the shared project.
OpenMuse — a personal agent with a computer of its own
I built the initial version solo in roughly a day, drafted the mascot and marketing, and worked on video direction. I continued developing the browser interactions, persistent sessions, and sourced comparison cards that make the agent's work visible and useful across mobile and web.
Case study · Example contribution: browser interactions
OpenDots — AI coworkers across text, calls, and Slack
I built the initial version in a day, then developed the workspace around specialist conversations, persistent documents, and visible computer work. My contributions include inline browser previews, file and terminal results, per-agent learning contexts, realtime call controls, and launch support.
Case study · Example contribution: initial workspace
OpenTag — agents in the channels where teams work
My work spans Channels SDK integration, a research backend, readable approvals for external writes, and developer onboarding for Slack and Teams.
Case study · Example contribution: readable approvals
OpenBot — agents people can direct and supervise
I extended the existing platform with action-specific approvals, queued corrections, repetition detection, and recovery from stalled runs. I also worked on creating skills and coworkers through conversation, alongside an internal deployment.
Case study · Example contribution: human approvals
OpenGenerativeUI — answers you can interact with
I worked on rich, agent-generated interfaces: interactive visualizations, charts, and 3D experiences. The project explores how agents can respond with useful interfaces alongside text.
- An AI product from idea to launch. Product direction, a working prototype, implementation, and a clear demo.
- Agents inside your product. Tool use, persistent context, human approvals, and interfaces that show what the agent is doing.
- Developer tools people can adopt. Reference applications, integrations, onboarding, and examples.
- Product and technical leadership. Turning an ambiguous opportunity into a focused roadmap and coordinating the work to deliver it.
- Folderbase — shared files and durable context for humans and agents working together.
- Chalk Agents — tools for testing, reviewing, and coordinating AI agent work.
More open-source experiments and starter kits
- Chalk Skills — reusable instructions for agent workflows.
- ClawUI Kit — a starter kit for CopilotKit and OpenClaw applications.
- ChiefClaw — an AI chief-of-staff experiment for freelancers.
- with-agent-spec — a Next.js and FastAPI starter with Agent Spec, AG-UI, and A2UI.
I've spoken on stage at Google events more than 10 times, including VibeCore at Google Mountain View, as well as developer conferences, university events, and team workshops. I share practical lessons from building AI products, leading teams, and working with agents.
I also founded the AI Cebu Community to help Filipinos develop practical AI skills.
Previously, I led product at Full Scale Ventures, helping ship two AI startups from zero to one, with work spanning MultiplAI and InterviewRoom.ai.
- LinkedIn Top Voice for Product Management, 2023.
- Named among the Top 100 Brightest Minds Under 30 in the Philippines by Stellar PH.
- Post Graduate Diploma in Artificial Intelligence and Machine Learning, Asian Institute of Management.
TypeScript, Python, React, Next.js, and Electron · CopilotKit, AG-UI, A2UI, LangGraph, and MCP · OpenAI, Anthropic, and Gemini · Docker, AWS, Google Cloud, and Firebase
I choose the stack around the product and the team.
Have an AI product to build or improve? Tell me about it →




