←
Transformer Lab330 posts
Transformer Lab avatar

Transformer Lab

@transformerlab

Building Primus, the world’s first fully autonomous AI researcher available to everyone. lab.cloud

◷ Joined December 2023▧ 63 Photos and videos
Transformer Lab avatar
Transformer Lab@transformerlab·Nov 12

Primus, our autonomous AI research scientist, is now running power semiconductor device research end-to-end in Professor Wai Tung Ng's Smart Power Integration and Semiconductor Devices Research Group at the University of Toronto. As a result, the first ever power semiconductor device designed end-to-end by an autonomous AI research system has been sent to fabrication. Primus runs end-to-end research inside the group’s device design workflow: • reviewing literature • planning and launching hundreds of linked simulations • analyzing the results and proposing the next round • delivering designs and writing up findings Why power devices? Power semiconductors convert the electricity behind EV chargers, MRI machines, and every chip in a data center. Demand is outrunning supply: the IEA expects data center electricity use to roughly double by 2030. Below are two early results and observations of Primus' impact on the group’s research progress.

Post attachment
Ali Asaria avatar
⟳ Transformer Lab retweeted
Ali Asaria@aliasaria·Nov 16

Nothing is holding us back anymore. At @transformerlab, we’ve officially started training large language models for public use. Current-gen compute is secured right here in Canada, and we’re already scaling training across multiple model families and modalities. If you’ve been waiting for a second company in Canada training at the frontier, it’s happening. Q1 is just the start. Tell us what you’re building or what use case you want solved, and let’s talk.

Ali Asaria avatar
⟳ Transformer Lab retweeted
Ali Asaria@aliasaria·Nov 13

"Now, how productive do you think an AI lab would be if it had 10,000 great researchers, each with this much compute, and each thinking 10 times as fast? You wouldn’t want to write a check to this neolab?"

The avalanche that hit math is coming for AI development. dwarkesh.com/p/rsi

44:26
Ali Asaria avatar
⟳ Transformer Lab retweeted
Ali Asaria@aliasaria·Nov 12

This is part of the work we've been waiting to show the world. Primus autonomously designed a 10 kV SiC diode end-to-end, cut an entire manufacturing step, and sent it to fabrication -- compressing ~100 hours of lab work into 1 hour of human review. Wild. In partnership with @MitacsCanada @UofT

As Professor Ng’s group continues to push Primus further into power device research, we are tackling hard open problems across fields like chip design, quantum computing, mathematics, physics, biotech, quantitative finance and more. If you’re interested in applying autonomous AI research capability into your company or research group, contact us. Read the full report: lab.cloud/news/uoft-power-se… Special thanks to @MitacsCanada for supporting our partnership.

Transformer Lab avatar
Transformer Lab@transformerlab·Nov 12

As Professor Ng’s group continues to push Primus further into power device research, we are tackling hard open problems across fields like chip design, quantum computing, mathematics, physics, biotech, quantitative finance and more. If you’re interested in applying autonomous AI research capability into your company or research group, contact us. Read the full report: lab.cloud/news/uoft-power-se… Special thanks to @MitacsCanada for supporting our partnership.

Transformer Lab avatar
Transformer Lab@transformerlab·Nov 12

Professor Ng's group estimates a study like the 10 kV silicon carbide diode would have taken about 100 hours of hands-on work without Primus. With Primus, the study required about 1 hour of hands on researcher time, to set the question and review the results and was completed in under 24 hours. Primus used about 10 hours of TCAD compute time. Primus ran simulations in parallel and handled deck setup, troubleshooting, monitoring, analysis and report drafting. "Primus is a transformative new approach to conducting semiconductor device research," said Wai Tung Ng, Professor at The Edward S. Rogers Sr. Department of Electrical & Computer Engineering, University of Toronto and Director of the Toronto Nanofabrication Centre. “My team is now exploring a far wider range of device structures and pursuing ideas we previously didn’t have capacity for. We’ve also become less reliant on external experts like materials engineers to inform parameter tuning approaches. Primus is materially accelerating our design process. It’s leading to discoveries the manual approach may miss and reducing the steep learning curve in training the next generation of semiconductor device engineers. We are excited to partner with Transformer Lab to evolve this discipline."

Post attachment
Transformer Lab avatar
Transformer Lab@transformerlab·Nov 12

Result 2: A 10 kV silicon carbide diode - Primus achieves one fewer manufacturing step while exceeding the published breakdown voltage AI data centres are moving toward taking power straight from the medium-voltage grid through solid-state transformers. Those run on ultra-high-voltage diodes, which fail at their edges first. Engineers etch a staircase of steps into the edge to spread the field. Every step costs another lithography and etch round. Primus started from a ultra high voltage diode design published less than a year ago that used four steps. Primus designed a three-step design, one less lithography and etch round that exceeded the published breakdown voltage of 13.61 kV in simulation, well above the 10 kV target, and documented the reasoning behind each change. The design is also forgiving. Its breakdown voltage stays within 0.3 kV even if an etch is off by 0.5 micrometres. This is the first design made end-to-end by an autonomous AI researcher sent to fabrication. Once verified in silicon, the design removes a full processing round from every wafer.

Post attachment
Transformer Lab avatar
Transformer Lab@transformerlab·Nov 12

Result 1: A high-voltage MOSFET: Primus achieves lower resistance at the same voltage. The MOSFET is the switch at the heart of a power converter. When it's on, its resistance turns power into heat. Add dopant to the drift region and resistance drops. Add too much and the device can no longer hold its rated voltage. Primus reviewed the high-voltage MOSFET literature, wrote the TCAD decks, ran and analyzed the campaign, then proposed follow-up rounds until the design converged. It worked the trade-off the way a seasoned device engineer would. The final design, a 200 V device with an optimized drift-region doping profile, reduced specific on-resistance down to 9.4 mΩ·cm². That’s in line with the optimum the reference paper predicts for this type of device. The group is preparing the full results for publication.

Transformer Lab avatar
Transformer Lab@transformerlab·Nov 3

We’re thrilled to welcome Ethan Thompson (@EthanT1452) to Transformer Lab as Lead Research Scientist in Quantum Science. Ethan joins us from a rich background, including the Perimeter Institute, where he worked on nonlocal QFT with John W. Moffat, professor emeritus of physics at the University of Toronto. Ethan will lead our work with physicists and quantum research groups applying Primus to some of the hardest problems in science. Read the full announcement: lab.cloud/news/ethan-thompso…

Post attachment
Transformer Lab avatar
Transformer Lab@transformerlab·Oct 23

Our co-founder @tonyfromkw is presenting at the MINT Workshop at #EMNLP2026 in Budapest next week. We show that FGD, a widely used metric for ranking co-speech gesture models, depends on settings that papers rarely report. Changing them changes which model comes out on top. The same settings also shift how closely FGD agrees with human ratings. This makes FGD scores difficult to compare across papers. Research was led by @tonyfromkw, @deepgandhi_07, and @aliasaria with support from Primus. Stop by and talk gesture evaluation with Tony 👋 🗓️ Thursday, Oct 29, 2:00-3:30pm CET 📍 MINT poster session, HUNGEXPO Budapest Congress & Exhibition Center @emnlpmeeting #EMNLP2026

Post attachment
Transformer Lab avatar
Transformer Lab@transformerlab·Oct 9

Two papers from our lab have been accepted to workshops at NeurIPS 2026 🎉 📍 Sydney: “Selecting Against Grammar” at LP4FM A common way to choose training data for a language model is to train small models on different mixes of data and keep the mix that scores best on question-answering benchmarks. We gave those same small models grammar tests. The mix that came first on question answering came 24th out of 25 on grammar. That mix is then used to train bigger, more expensive models, with no warning that its grammar problems could carry over. 📍Atlanta: “One Requirement, Many Specifications” at VeriCodeGen Some tools can mathematically prove that code is correct, but only by checking it against a written rulebook that spells out exactly what the code should do. Writing these rulebooks takes experts, so researchers are increasingly having AI write them. If the AI gets the rulebook wrong, the code is proven correct against the wrong rules, and the proof makes it look trustworthy anyway. We asked a coding AI to write the rulebook for the same task 16 times, to see if it gives the same answer each time. On half the tasks we could test, it wrote rulebooks that contradict each other: for the same input, one says an answer is right and another says it's wrong. Which rules your code gets proven against comes down to luck. The work on both papers was led by @deepgandhi_07, @tonyfromkw and @aliasaria. More details coming soon! 🚀 #NeurIPS2026 @NeurIPSConf

Ali Asaria avatar
⟳ Transformer Lab retweeted
Ali Asaria@aliasaria·Sep 25

Here is a link to the paper we published about how Primus Society works zenodo.org/records/22922326 Inside is a lot of the thought we put into how the society is organized and why we made specific decisions. A big focus for us is: "Give agents a purpose, not a goal"

Post attachment
Transformer Lab avatar
Transformer Lab@transformerlab·Sep 24

Introducing Primus Society: an agent society of thousands of autonomous AI researchers. We believe this is the path to scaling AI research safely at unprecedented scale safely and productively. Learn more about Primus Society, the research it produced and browse the virtual city yourself: lab.cloud/society/

We now run the largest frontier lab composed entirely of autonomous AI researchers. Introducing Primus Society: a society of agents composed of thousands of researchers working under structured institutions designed to solve the world’s toughest problems. We believe this is how AI research should run at scale, safely and productively: an entire society of agents with institutions and purpose. One discovery we can already share is that the society has discovered a novel result which improves model training by 30%. Primus Society was inspired by the structures that have organized science for centuries and stress-tested against what’s known about how populations of AI agents fail. Everything is observable. You can open the virtual city in a browser and read what any researcher is working on. Learn more about Primus Society here: lab.cloud/society The biggest opening in the AI race is running the largest well-governed organization of AI researchers in the world. We’re building the institutions that let a million AI scientists safely tackle the world's toughest problems in AI and beyond. And we’re doing it right here in Canada.

0:24
Transformer Lab avatar
Transformer Lab@transformerlab·Sep 22

Physics is now a specialization you can select in Primus. It's available to everyone starting today. It's built for the work physicists do every day but rarely have time to check line by line: deriving an equation of motion, working out an effective Hamiltonian, evaluating an integral, taking a limit and confirming it behaves. Primus doesn't just do that work. It checks the answer. In Lean it uses Physlib, the physics library covering classical mechanics, electromagnetism, relativity, statistical mechanics and QFT. Outside Lean it recomputes the result in sympy, scipy and mpmath. When a result exists only as a number, PSLQ proposes a closed form and Primus tests it. Every result in the writeup says how it was established: machine-checked, verified by computation, or heuristic. All calculations are in the activity log, so you can trace claims back to the steps that produced them. Give it a try: lab.cloud

1:15
Transformer Lab avatar
Transformer Lab@transformerlab·Sep 18

We launched Primus for Math in partnership with Maplesoft. It's a mode of Primus built specifically for math and engineering using the Maple engine. It’s available in beta to all Primus users starting today. Learn more: lab.cloud/news/primus-for-ma…

Big update: Primus for Math is live today. It’s an autonomous AI research scientist specifically for mathematics and engineering. We built it in partnership with Maplesoft to make frontier AI mathematics research systems accessible to everyone, not just a few frontier closed companies. Primus for Math is a true partner for math researchers and engineers. Give it a problem and it reads the field, works through the analysis in Maple's math engine, simulates the system, proves what it can and shows its work at every step for you to understand and build on. It also changes who gets to do this kind of work. Now a physicist with a math question, an engineer working outside their field, a two-person lab can do the work themselves as Primus brings the field's literature, methods and rigor to the project. Maplesoft has been an ideal partner to collaborate with. During the beta, they’re generously offering Maple access for every Primus for Math user at no additional cost. And we're offering free credits to Primus if you sign up now.

0:32