Know before you synthesize

Researcher working at a laboratory bench

Our mission

Heuresia is here to take the biggest reason drugs die out of the equation, by building an engine that predicts whether a molecule will be toxic from its structure alone, before anyone has made it.

Our team of deep-learning engineers and university toxicologists builds predictive models that are grounded in physics and real assay data, carry honest uncertainty on every call, and are judged on public benchmarks frozen before we train.

Each cycle of results sharpens the engine. We are proving it on compute first; the drugs come next. Every step brings us closer to our mission: a future where no drug fails on safety.

Part of

altair

Venture Capitalaltalab

Founder

Founder & CEO

I started Heuresia as a medical student at the Medical University of Warsaw, watching what machine learning was beginning to do everywhere except the field I was training for. Drugs still die late, on safety, after years of work and money. That looked like a problem worth a whole company, and I decided to take it on.

I chose safety because it is the highest-value call in any programme. Toxicity kills more drug candidates than anything else, and it is usually found late, after the money is spent. I want us to be the best in the world at that one prediction.

My job is all of it: keeping the vision shared, the team productive and the engine honest. I bring the medicine, the education and the faculty connections that turn a benchmark result into something a toxicologist will stand behind. If you back deep tech where machine learning meets biology, invest in this journey.

A line I live by

“If your dream still lives, if that castle gleams just as brightly in your eyes, then it is your obligation to lay the stones that surround you now.”

Void, in Berserk by Kentaro Miura

Co-founder

I’m a Finance student who has always been interested in what happens when technology meets capital. When I joined Heuresia, I saw a chance to apply that mindset to an industry where both are pushed to their limits. Drug development can take years and millions in capital, only for a promising candidate to fail late. That looked like a problem worth building around.

I chose Heuresia because better prediction can change where capital gets deployed in drug development. If we can identify failure earlier, we can stop wasting years and resources on candidates that were never going to make it.

My job is all of it: finance, fundraising, strategy and operations. I work on turning the engine into a company, finding the capital, building the partnerships and making the decisions that keep us moving. I’m still a student. I just didn’t see a reason to wait.

A line I live by

“You can’t win if you don’t play.”

01 / 02

Know before you synthesize

Drugs fail on safety. We predict it first.

Mark Sobolewski