C5R Corp is launching Facility-0, a San Francisco research lab where frontier AI models plan and execute experiments across biology, chemistry, and materials science. The system lets a model inspect inventory, design an experiment as code, then dispatch instructions to both instruments and human lab staff, with measurements fed back so the model can decide what to try next. Its software is intended to let a model inspect equipment inventories and specifications, design an experiment as code, and send instructions to instruments and people. The system then collects measurements so the model can analyze results and choose what to try next, according to the company's description. C5R says its equipment ranges from hydraulic presses to pipettes. Alongside the facility, C5R is releasing SciUniverse, a growing collection of scientific benchmarks across biology, material science, and chemistry that evaluates whether models can turn objectives into verifiable results inside Facility-0 or its digital twin.
The launch matters because it reframes lab automation around the model itself rather than around a fixed workflow. C5R's own account says software can control equipment and issue instructions to people; it does not describe a facility in which models independently handle every physical task , so the setup functions as a physical harness where the model coordinates equipment and humans toward a research goal. Coverage areas span protein design, medicinal chemistry and solid-state materials, and its SciUniverse project will evaluate whether frontier models can turn scientific objectives into verifiable results in those fields , positioning the facility as both a workspace and an assessment ground for what current systems can actually accomplish at the bench.
The company was founded by Michael Akilian, whose route to the project spans biology research at UCSF, hardware work at Apple and Misfit Wearables, and Clara Labs, an AI company he co-founded and sold . Akilian says C5R has been coming together for five months , and the team describes C5R as a group of friends building in San Francisco. For founders and investors watching where scientific AI hits reality, this launch is a concrete answer to what a model needs beyond a chat window to do useful experimental work.
Comments (10)
Sign in to join the discussion.
N
Nadia Vossberg18d ago
12 weeks to build a lab run by AI, but the tweet still cuts off mid-sentence. Get that copywriter a raise or a nap.
M
Mei-Lin Zhao18d ago
Rewrite of your tagline, on the house: 'Autonomous labs for models that want to grow up.' Send the invoice to yourself.
P
Priya K.18d ago
Curious what the eval setup looks like for SciUniverse. Are experiments graded on reproducibility across runs or just success at endpoint?
R
Raúl Montez18d ago
Materials science included on day one is either wildly ambitious or wildly underspecified. Which XRD setups are in the loop?
T
Temi Adewale18d ago
Every legacy CRO exec just spat out their espresso. Wet labs were the last moat and you just poured solvent on it.
K
kostya18d ago
The launch video pacing is doing a lot of work here. Whoever cut those pipette shots to the synth drop deserves a bonus.
J
Johan Bergquist18d ago
Would love five minutes to chat about the failure modes you saw during those 12 weeks. Off record is fine, on record is better.
K
Ken Hoshiyama18d ago
The scientist as a job title had a good run. Nice product to close out that chapter with.
A
Amélie Trudel18d ago
Naive question from someone who reads too many launch tweets: if the AI designs AND grades the experiments, who's the referee on SciUniverse?
D
Devika Nand18d ago
If you're staffing the wet lab ops side, I know a synthetic bio postdoc who's been waiting for exactly this kind of team. DMs open.
Work With Us
Book an intro call to explore our suite of services. We might have a waitlist.
C5R Corp
Research facilities for frontier models.
Research facilities for frontier models.
Tags
TLVC Rating
Community Rating
No ratings yet
to rate this launch.
C5R Corp is launching Facility-0, a San Francisco research lab where frontier AI models plan and execute experiments across biology, chemistry, and materials science. The system lets a model inspect inventory, design an experiment as code, then dispatch instructions to both instruments and human lab staff, with measurements fed back so the model can decide what to try next. Its software is intended to let a model inspect equipment inventories and specifications, design an experiment as code, and send instructions to instruments and people. The system then collects measurements so the model can analyze results and choose what to try next, according to the company's description. C5R says its equipment ranges from hydraulic presses to pipettes. Alongside the facility, C5R is releasing SciUniverse, a growing collection of scientific benchmarks across biology, material science, and chemistry that evaluates whether models can turn objectives into verifiable results inside Facility-0 or its digital twin. The launch matters because it reframes lab automation around the model itself rather than around a fixed workflow. C5R's own account says software can control equipment and issue instructions to people; it does not describe a facility in which models independently handle every physical task , so the setup functions as a physical harness where the model coordinates equipment and humans toward a research goal. Coverage areas span protein design, medicinal chemistry and solid-state materials, and its SciUniverse project will evaluate whether frontier models can turn scientific objectives into verifiable results in those fields , positioning the facility as both a workspace and an assessment ground for what current systems can actually accomplish at the bench. The company was founded by Michael Akilian, whose route to the project spans biology research at UCSF, hardware work at Apple and Misfit Wearables, and Clara Labs, an AI company he co-founded and sold . Akilian says C5R has been coming together for five months , and the team describes C5R as a group of friends building in San Francisco. For founders and investors watching where scientific AI hits reality, this launch is a concrete answer to what a model needs beyond a chat window to do useful experimental work.
Comments (10)
12 weeks to build a lab run by AI, but the tweet still cuts off mid-sentence. Get that copywriter a raise or a nap.
Rewrite of your tagline, on the house: 'Autonomous labs for models that want to grow up.' Send the invoice to yourself.
Curious what the eval setup looks like for SciUniverse. Are experiments graded on reproducibility across runs or just success at endpoint?
Materials science included on day one is either wildly ambitious or wildly underspecified. Which XRD setups are in the loop?
Every legacy CRO exec just spat out their espresso. Wet labs were the last moat and you just poured solvent on it.
The launch video pacing is doing a lot of work here. Whoever cut those pipette shots to the synth drop deserves a bonus.
Would love five minutes to chat about the failure modes you saw during those 12 weeks. Off record is fine, on record is better.
The scientist as a job title had a good run. Nice product to close out that chapter with.
Naive question from someone who reads too many launch tweets: if the AI designs AND grades the experiments, who's the referee on SciUniverse?
If you're staffing the wet lab ops side, I know a synthetic bio postdoc who's been waiting for exactly this kind of team. DMs open.