Post 06Reframe

Open-Sourcing My Biology

William Kasel·1 min read

Publishing genome, bloodwork, protocols, and AI tools on GitHub

The 30-Second Version

Duration: 40 seconds Compliance: No compound names needed in this one - it's about the concept and the tools.

"I'm doing something that might be either really smart or really stupid.

I'm publishing my entire genome, my bloodwork, and every protocol I'm running - publicly. On GitHub. For free.

Plus the AI tools I built to analyze all of it.

Here's why: the health optimization space has a transparency problem. Everyone shares results but hides the work.

I think it changes if people can see real data, from real people, making real decisions. With the genetic reasoning behind every single choice.

I'm a tech founder. I build in public. This time the product is my own biology.

I work with a medical team. Everything linked in bio. Fork the code. Use it on your own data. Let's go."

Last line (quotable): "Building in public - except the product is my own biology."

01

I'm making my health data public.

My genome. My bloodwork. My protocols. The AI agents I built to analyze all of it.

Everything goes on GitHub. Here's why. 🧵

02

The health optimization space has a transparency problem.

Influencers share results but hide protocols. Companies sell tests but don't explain the data. Doctors gatekeep information behind appointments.

I think the whole game changes if people can see real data from real people making real decisions.

03

What I'm publishing:

  • 178 analyzed genetic variants with full interpretations
  • Longitudinal bloodwork (100+ markers tracked over time)
  • Full supplement and protocol stack with genetic reasoning for each item
  • The AI agents I built to do the analysis
  • The research library behind every decision
04

Why this matters beyond me:

If 100 people published their genetics + protocols + outcomes, we'd have a better dataset than most clinical trials.

N=1 is anecdotal. N=1,000 with genetic data? That's research.

I want to be one of the first N.

05

The AI tools are the part I'm most excited about.

I built agents that:

  • Analyze raw 23andMe data against clinical databases
  • Cross-reference genetic variants with supplement protocols
  • Flag drug-gene and supplement-supplement interactions
  • Mine Reddit/TikTok for trending health questions

All open source. All on GitHub.

06

I'm a tech founder, not a doctor. I build things.

This is me building in public - except the product is my own biology.

Follow along. Fork the code. Use it on your own data. Let's make health optimization open source.

[Links to repos]

This is what I put in The Manual every week.

Subscribe to The Manual →

More Posts