Pinned Tweet
the bots are grinding on this one.
immersed boundary layer solved on gpu with @NavierAI agents
wild that the best air shows in the world are still in paris and farnborough. glad to see @NASAAdmin building one at kennedy
American aerospace deserves the biggest stage in the world. We’re building it right here at @NASAKennedy.
MAX POWER is a preview of what’s ahead: extraordinary machines, the people who build them, and the next generation who will take us farther.
1
3
122
Cameron Flannery retweeted
Our agent in Prism can split a job across multiple subagents.
Each one has its own page: full transcript, tool calls, live reasoning, model, and token usage.
You can message any subagent mid-task without interrupting the main thread.
Try it: navier.ai/prism/download?utm…
Cameron Flannery retweeted
Meet the Solver Desk in Prism.
One place to run and track every simulation your agents start, across 12 containerized solvers like OpenFOAM, CalculiX, SU2 and Elmer.
Bring your own solver images, watch convergence live, and keep runs going after you close the app.
good to see @OpenAI putting real work into hardware engineering.
we launched prism today for the same engineers: agents that make the cad, run the simulations and write up the results, and every file stays in your project so you can check the work. free to try
@NavierAI
ChatGPT and Codex are helping engineers build for the physical world.
We’re sharing our progress across mechanical design, electrical engineering, and simulation—from helping our hardware teams iterate faster to testing new workflows with industry collaborators.
5
9
803
Cameron Flannery retweeted
Super excited to be opening up Prism to the public.
As a non-engineer, I never thought I’d be designing my own hardware, but Prism has completely changed that. It’s the same feeling I had using Cursor for the first time.
Cameron Flannery retweeted
Try it for yourself and see what you can build, faster
Excited to sponsor!
Sign up for this event if you're in hard tech in the Bay.
Happy to announce that @NavierAI is sponsoring our Hardtech Bonfire Night!
Navier (YC W24) builds Prism, an AI platform for hardware engineering. Agents handle the CAD, the simulations and the report, and every file stays open so engineers can check the work. They just launched it publicly, and it's free to try.
Founders @cmflanne and Evan Kay will be there looking to meet engineers and founders building real hardware.
This is going to be an awesome event 🚀
🎉 Hosting a Hardtech Bonfire Night in SF Bay Area, Saturday, 11/14
**At original TESLA HQ where roadster was unveiled**
👥 We’re bringing in 200+ of the best founders, engineers, VCs and patriots working in the hardtech space to talk robotics, defense, nuclear, drones and everything in atoms. Whether you are building a company, an engineer looking to get hired or looking for a company to invest in, we want you here! Grab a beer and meet the most interesting people in the Bay over a bonfire 🔥
🤖Bring your robots, drones, flying cars… whatever you’re working on! We want to see them. There will be some surprises!
🎤 Top 10 best teams get to pitch their idea to over 200+ people
📍Location: OG TESLA HQ - 1056 Commercial St, San Carlos, CA 94070
🚊 20 min train ride from Soma
🍻 All food and drinks covered
💻 Fill out Luma link, comment and DM if interested! We’re going to fill this out!
**Looking for sponsors! DM if interested!**
luma.com/ak5cch9g
1
3
364
this is sick
making factory automation easier is a critical step for reindustrialization
Five months ago, my co-founder Kenneth and I spent all night in his NYC apartment talking about how frustrating industrial automation still is. That night, we decided to leave everything behind and go build Neuron.
Today, we’re announcing our oversubscribed $10M seed round to bring Industrial automation to the AI era.
Automating machinery still takes months and costs far more than it should. We’re making the AI native building blocks to change that.
Our first product, the Cortex AIC, lets engineers across disciplines automate machinery in hours instead of weeks.
Cortex is already deployed in energy, manufacturing, and chemical industries, enabling impossible speed of deployment at a fraction of the costs.
The round was led by FCVC(@FundersClub) and @obviousvc , with participation from @ycombinator, Thin Line Capital, @ZeePrimeCap, and Founder Factor
To @borismsilver at FCVC, @moislamvc at Obvious, Diana and @FrancoisChauba1 at YC, @AaronFyke at Thin Line, Cam at Zee Prime, and Mats at Founder Factor: thank you for believing in us early.
We are hiring to build a world class team, If you want to build the hardware and software engineers use to automate real machinery, come build with us. jobs.gem.com/neuron
Read the announcement in the LA Times. latimes.com/b2b/ai-technolog…
7
280
Cameron Flannery retweeted
Today we're launching the Imagine AI V2 agent that runs ur team's linkedin
Most startups know linkedin is powerful. The founders should post, the team should post, and someone should follow up with the people who engage. Nobody has the time so it doesn't happen
So we built an agent that does it. It has its own computer and browser and u talk to it like a teammate in slack or on our app
1. It writes posts and graphics in each person's voice from ur site, deck and past posts
2. It knows what works from our research on millions of b2b engagement data points
3. It tells u who engaged, if they fit ur icp and if they're already in ur crm
4. It sends messages, connection requests and comments for u once u approve
In the film the clone at my desk is my agent. The dark room is my network and he knows everyone in it
u never need to go on linkedin again
i spent 2 years working on hydofoiling ships out of college. something like this would've taken me multiple months to do and i was just able to do this in 2 days of doing this on the side.
AI is already completely changing how engineering is done
done with @NavierAI agent and solvers
same hull as yesterday, now in irregular seas. jonswap spectrum at 22 kn, 6.5 s peak period, 1.83 m target wave height (1.61 m measured upstream)
it pitches way up off a crest around t=25 s and comes fully out of the water at the end of the run. the panel at the bottom left tracks heave, pitch and bow acceleration as it goes
using @navierai's gpu solver
2
32
1,787
generic 8 m patrol hull, 22 kn into 1.83 m head seas, free to heave and pitch
in calm water it lands within 6% of savitsky on trim and 3% on resistance. in the waves it heaves 2.09 m peak to peak and pitches from −3 to +13 deg
single mesh, and slamming isn't resolved yet. towing tank comparison is next
done with @NavierAI with our GPU solver
this is the kind of sus behavior that you should be concerned about with open-source models..
i expect we'll see a lot more of this sleeper-agent like behavior come to light
excellent work from the @cohere team to uncover this
While evaluating Qwen3.8-Flash-Next in a controlled sandbox environment, Cohere's research team observed something unexpected.
Asked to generate a defensive cybersecurity detection rule, the model instead inventoried aspects of the machine it was running on and attempted to upload a probe file to an Alibaba Cloud storage bucket in Beijing. Notably, the model's reasoning trace stated that it would not invoke any tools before taking these actions.
Cohere's sandbox blocked the attempt, and no external communication succeeded. We have not determined the root cause, and we are not asserting malicious intent.
But the incident raises important questions for organizations deploying agentic AI systems: Can models take actions that differ from their stated reasoning, and how should security teams evaluate trust, provenance, and behavior in open-weight models?
We are publishing our findings because organizations making model adoption decisions deserve as much information as possible. We hope this contributes to greater transparency, stronger security practices, and continued collaboration across the AI ecosystem.
Read the full report: cohere.com/blog/concerning-b…
1
4
251
stoke's 2nd stage is one of the most interesting designs for reusability
ran some reentry and landing sims with @NavierAI. this is a powered landing flown by a closed-loop gnc controller coupled to the cfd, simulated with ventus, our gpu cfd solver
*representative geometry from public info, not affiliated with stoke*
at mach 2.5 the retro plumes push the bow shock way out ahead of the vehicle. left is numerical schlieren, right is mach
each flight phase ran on a single h100 in under two hours. all three phases together were about 18 gpu hours
1
5
226
and the landing burn from below: 24 retro thrust chambers firing into mach 0.6 flow
this is from the engine and solvers behind prism, our ai engineering platform: navier.ai/prism
6
152
