Building the AI layer for manufacturing systems at @fdotinc | Built by @hasnainbaig06, @jahenderson05, @ryanmirdain
Joined June 2026
- Tweets95
- Following7
- Followers90
- Likes329
Mirdain Systems retweeted
Working to re-industrialize America has never been more important. Over 90% of 3D printers today are built and sold by Chinese companies, with no single big US company to take Bambu or Creality.
An American 3D printer should be and will be the most reliable, high quality, and best option on the market. That's why are building our own printers, with our end to end AI software directly integrated into the printer.
Mirdain Systems retweeted
This is going to be the biggest physical AI hackathon of the year in SF
We're hosting a huge Physical AI hackathon.
We’ve run robotics hackathons before but this one is different:
Bimanual Metal Arm and Maker Arm Stations
Fine Tune Dream-1, this is not another ACT competition
New Hill Climb leaderboard format
Two tracks: push your robot up the live eval leaderboard. Or build the creative, most useful robot application you can think of.
Partnering with @dreamlabs_ai @livekit @huggingface @nvidia @_lightcompany @mirdainsystems
November 7–8. San Francisco.
Co-hosted with Founders Inc @fdotinc
Come build with us: luma.com/pah
can’t wait!!
We're hosting a huge Physical AI hackathon.
We’ve run robotics hackathons before but this one is different:
Bimanual Metal Arm and Maker Arm Stations
Fine Tune Dream-1, this is not another ACT competition
New Hill Climb leaderboard format
Two tracks: push your robot up the live eval leaderboard. Or build the creative, most useful robot application you can think of.
Partnering with @dreamlabs_ai @livekit @huggingface @nvidia @_lightcompany @mirdainsystems
November 7–8. San Francisco.
Co-hosted with Founders Inc @fdotinc
Come build with us: luma.com/pah
Mirdain Systems retweeted
We're hosting a huge Physical AI hackathon.
We’ve run robotics hackathons before but this one is different:
Bimanual Metal Arm and Maker Arm Stations
Fine Tune Dream-1, this is not another ACT competition
New Hill Climb leaderboard format
Two tracks: push your robot up the live eval leaderboard. Or build the creative, most useful robot application you can think of.
Partnering with @dreamlabs_ai @livekit @huggingface @nvidia @_lightcompany @mirdainsystems
November 7–8. San Francisco.
Co-hosted with Founders Inc @fdotinc
Come build with us: luma.com/pah
Mirdain Systems retweeted
We killed 3D printer fleet management companies.
Mirdain already spans print preparation, AI defect detection, and printer control across any printer
Your entire category is our starting point.
A large part of 3D printing has been stuck in the past, and machines on the market don’t compare to the technology we have today. We are solving that by making the machines smarter, and think for themselves instead of dealing in absolutes.
What we are building:
- Vision based defect detection
- Real time changes to production based on defects
- Real time g-code intervention
- Fleet management
- Printer/bed/nozzle wear
- AI assisted slicing
- "texting" your printer to make live changes and edits
Building toward fully autonomous 3D printing
Mirdain Systems retweeted
What we are building:
- Vision based defect detection
- Real time changes to production based on defects
- Real time g-code intervention
- Fleet management
- Printer/bed/nozzle wear
- AI assisted slicing
- "texting" your printer to make live changes and edits
Building toward fully autonomous 3D printing
Mirdain Systems retweeted
Absolutely hate how operation heavy the whole process is, and frankly, it’s outdated. There should be no real reason to spend so long in the slicer, or to even sit by ur print to ensure it’s coming out well. Quality should be something guaranteed
Mirdain Systems retweeted
We are heading back to where it all started this Friday to talk more about what we are working on!
If anyone in the Santa Clara area wants to learn more about what we are working on, or just wants to chat, come say hi!
Mirdain Systems retweeted
Box is 3D printed btw, wonder if anyone could tell
We just built this. This little box lets me fully manage and auto slices based on whats most optimal, letting me make the most of all my printers. Once I clear a part off my bed, it knows exactly what it needs to print next, and starts it up as soon as it sees the printer is ready.
Next step is to be able to text my printers
Mirdain Systems retweeted
We just built this. This little box lets me fully manage and auto slices based on whats most optimal, letting me make the most of all my printers. Once I clear a part off my bed, it knows exactly what it needs to print next, and starts it up as soon as it sees the printer is ready.
Next step is to be able to text my printers
Mirdain Systems retweeted
+40% over vision only on contact rich tasks.
great work from @LerrelPinto's lab 👏🏼
touch isn't optional for manipulation.
cheap BOM + @mirdainsystems for 3D prints, the hard part is no longer the sensor.
it's making touch hold up on a factory line at 99%+ yield, every shift.
Open-source magnetic tactile sensor for $5! 🧲
Researchers from New York University introduced a magnetic tactile sensor that's low-cost, and easy to fabricate, democratizing tactile sensing for robotics.
Operating in unstructured environments like homes and offices requires robots to sense forces during physical interaction. Yet the lack of a versatile, accessible tactile sensor has led to fragmented solutions and often force-unaware, sensorless approaches.
Building an eFlesh sensor requires four components: a hobbyist 3D printer, off-the-shelf magnets (less than $5), a CAD model, and a magnetometer circuit board.
The sensor is 3D printed with magnets embedded in the middle layer. Based on chosen mechanical properties, magnets displace in response to contact forces, measured by a magnetometer underneath.
An open-source design tool converts simple OBJ/STL files into 3D-printable STLs. This enables application-specific sensors for robot hands, grippers, quadruped feet, and more.
Slip detection generalizes to unseen objects with 95% accuracy. Visual-tactile control policies improve manipulation by 40% over vision-only baselines, achieving 90% success on precise tasks like plug insertion and credit card swiping.
All design files, code, trained models, and conversion tools are openly available.
🔗 Project page: e-flesh.com
~~
♻️ Join the weekly robotics newsletter, and never miss any news → ziegler.substack.com
Mirdain Systems retweeted
AI models just keep getting better and more efficient and autonomous, but bringing that into more physical sectors like manufacturing is a much bigger task. I think soon enough we will be able to fully rely on models to think and make decisions the same way we do, making it the best time to be building a connective layer between for these physical systems.
Soon enough we will be able to offload operating and programming CNCs, or letting them rewrite gcode in real time to get something optimal in additive manufacturing.