Founder-engineer building AI systems for the physical world. PromptPCB, AutoTQ, robotics, PCB design, firmware, ML, and medical devices.
Gainesville
Joined July 2019
- Tweets435
- Following95
- Followers148
- Likes90
Pinned Tweet
I’m rebuilding this account around what I actually do now:
I build AI + robotics systems for the physical world.
PCB design. Embedded C. Robotics control. ML/computer vision. Mobile apps. Medical-device engineering. Mechanical prototypes. The messy stuff where software has to survive contact with hardware.
I’ve founded and built projects across pathology imaging, automated agriculture, medical devices, AI training tools, PCB automation, and robotics. I’m currently building PromptPCB and leading engineering on AutoTQ.
This account is going to be build logs, technical notes, failures, fixes, prototypes, AI hardware workflows, and lessons learned the hard way.
No guru content. No fake founder theater.
Just engineering, product building, and the grind of making difficult systems work.
8
745
Fable just decided to calibrate my robot camera using the angle my shadow created during a recording...
3
80
Relatively basic sim but produced with no external libraries with integrated physics as well as accurate harmonic drive motion
Nothing brand new happening here but trying to push the limits of the models ability to create and organize new tools for itself.
The only prompt given was "make a physics accurate robot arm" it decided fully what tools and framework needed to be created from scratch and reorganized itself merging duplicate tools and code and building on what it had produced prior
Currently working on building the "skill tree" for this agent
2
92
Put opus 5.5 in a sandbox with only access to python and system level programs. No external libraries and nothing beyond that. Gave it a system level prompt telling to organize itself, its folders, and create tools to achieve the users requests.
First it created itself curl
Second it found the ESP32 datasheet and made a PDF parser
Third it located the footprints and STEP files for these parts
Fourth it create a full processor to validate the STEP files and produce and render the mesh
Fifth created a reusable 3d engine it could render the parts in
Sixth Made them look very nice and didn't stop until they had correct textures
It did all this while keeping everything very organized and continually modify and rewriting the tools it made earlier in order to be more modular and minimize duplciate code
they actually made it
Meet Mistral Large 4, aka Le Chonk.
• 1T parameters, natively multimodal. 49B active.
It is the best open weights model from US or Europe on aggregated benchmarks.
• State-of-the-art on critical workloads, including cyber defense, manufacturing and finance and it surpasses closed frontier models on visual grounding.
• Forged in Europe end-to-end and is deployable from Europe via our own Mistral Cloud infrastructure.
• Available to all via API today. Working with cybersecurity partners privately.
Open weights release end of October.
24
jev for stonks.
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution
36
