I tried to build a VC-backed startup. Failed to find something venture-scale that I deeply cared about. But it taught me more about myself than any other experience. It was definitely worth it.

I spent 9 months building LLM-powered apps to help cancer patients navigate clinical care—chatbots with EHR integrations, appointment prep tools, and iOS apps (that Apple kept rejecting). I found real use cases and built working prototypes, but I discovered that LLMs aren’t ready for safe clinical use by patients. The core structural issues—context building, grounding, and usefulness—won’t be solved by the next model release.

A long and and imperfect post about my brush with mortality, music making during my career break and my new resolution for 2025.

I went through the journey of fine-tuning two small LLMs: Llama 3.1-8B-Instruct and GPT-3.5-Turbo on a fun dataset of GPT-4o generated conversations in the style of Larry David. All the code for this project and evaluation results are openly accessible.

I built AI Dialer, a full stack Python app for interruptible and near-human quality AI phone calls by stitching LLMs, speech understanding tools, text-to-speech models, and Twilio’s phone API. This post describes the process for building this project as well as challenges and opportunities discovered in the process.