Personal AI Agent with Voice

I built a personal assistant around a single agent loop: the model is offered a set of tools, decides which to call and in what order, and keeps going until it has an answer. Each capability is one function plus its description, so the assistant grows by adding tools rather than by adding branches.

It started with web search through Tavily, then the ability to open a page and have a cheaper model read and score it. Persistent memory came next through mem0, so it recognises me across sessions instead of starting cold.

The job search is where those pieces add up. It reads an uploaded resume, searches for live openings, scores each one against that background, and emails the shortlist through Resend. Because memory was already in place, it remembers the resume too.

The last two steps removed the screen. A Twilio number connects to an ElevenLabs voice agent, and four voice tools call the same endpoints the web app uses. I can phone the assistant, ask it to run the job search, and hang up while it works and emails the results. Chat and phone are one assistant reached two ways, sharing one memory.