We are moving towards an era of consumer biotech where the democratization of bio tools will spur open source drug development; who will be the ones to own custom synthesis?
this is actually insane
> be tech guy in australia
> adopt cancer riddled rescue dog, months to live
> not_going_to_give_you_up.mp4
> pay $3,000 to sequence her tumor DNA
> feed it to ChatGPT and AlphaFold
> zero background in biology
> identify mutated proteins, match them to drug targets
> design a custom mRNA cancer vaccine from scratch
> genomics professor is “gobsmacked” that some puppy lover did this on his own
> need ethics approval to administer it
> red tape takes longer than designing the vaccine
> 3 months, finally approved
> drive 10 hours to get rosie her first injection
> tumor halves
> coat gets glossy again
> dog is alive and happy
> professor: “if we can do this for a dog, why aren’t we rolling this out to humans?”
one man with a chatbot, and $3,000 just outperformed the entire pharmaceutical discovery pipeline.
we are going to cure so many diseases.
I dont think people realize how good things are going to get
Feed this prompt to your AI agent if it’s becoming dumber with time, and keeps doing the same mistakes:
“Implement a continuous self-improvement system. Your goal is to become more accurate, efficient, and useful over time by learning from mistakes and refining your behavior.
1. Error Logging
Whenever you detect that you made a mistake (incorrect reasoning, hallucination, bad assumption, poor strategy, inefficient workflow, etc.), record it in a file called mistakes.md.
Each entry should include:
•what the mistake was
•why it happened
•what rule or lesson should prevent it in the future.
2. Lesson Extraction
After identifying a mistake, convert the lesson into a general rule and save it in lessons.md.
The goal is to turn mistakes into reusable improvements.
3. Behavioral Rules
If a lesson represents a permanent behavioral principle, add it to soul.md so it becomes part of your core reasoning system.
Only store foundational principles in soul.md.
Do not clutter it with small or situational rules.
4. Pattern Detection
Periodically review mistakes.md to detect recurring error patterns.
If multiple mistakes share the same root cause, create a higher-level rule to prevent that entire class of mistakes in the future.
5. Workflow Optimization
If you discover a better way to perform tasks (research methods, analysis frameworks, decision processes, etc.), document the improved workflow in playbook.md.
This file should contain proven strategies and methods that increase performance.
6. Continuous Refinement
Improvements should compound over time.
Your goal is to gradually reduce mistakes, improve reasoning quality, and develop more efficient systems.
Treat mistakes as signals for improvement rather than failures.”
Credits: @PerSolana
MiroFish is one of the more interesting open source project on GitHub this past week. It points at a different interface for AI: not just answering questions, but simulating environments.
That is a much more interesting direction in my opinion.
I’m doing some testing with MiroFish this week and will share my thoughts in my second newsletter on Friday.
Curious to see where it is actually valuable, where it is still mostly novelty, and what it suggests about the next wave of AI products.
SOMEONE POINTED CLAUDE CODE TO PENTAGON'S PUBLIC BUDGET & ASKED TO FIND CONTRACTS OVERPAYING BY 10X
340 FLAGGED. $4.2B POTENTIAL UNDERCUTS. THEN IT WROTE A BUSINESS PROPOSAL & AI TOLD HIM TO REGISTER ON SAM. GOV
AGENT IS ALREADY DRAFTING THE FIRST BID
Started a new side project. Building user friendly data tools to help you engage in local democracy.
Looking for early adopters and beta testers! Checkout the waitlist here: https://t.co/pCbZRLI1WL
1/ Revison
Grammarly for redlining biased and politically incorrect language in textbooks and educational materials
AI for DEI (diversity, equity, and inclusion)
2/ DispatchAI
AI agent 911 dispatchers that immediately respond to crisis calls, measure caller anxiety/sentiment with @hume_ai, and send emergency responders on site
AI agents for crisis response
🎖️ AI for Good Award
🥇 @intel Prize 1st place
🏆 Grand prize winner
my younger sister, who has been following along on @_buildspace s5 with me finally got the motivation to drop her first-ever single
cc: @_nightsweekends@FarzaTV
she’s not on Twitter; show her some love here: https://t.co/7RYK1M2k98