Increasing Humanity's Healthspan
We develop novel methods, user-friendly software, and interpretable AI models to design biologic medicines.
Where we start
What does it take to make a viable therapeutic?
48% and 55%
of phase II and phase III failures came from a lack of efficacy.
25%
of phase II failures came from safety, the next most common reason.
Our services
We are addressing the pharmaceutical industry's biggest pain points:
Sources: Sun et al., 2022, Acta Pharmaceutica Sinica B 12(7), doi:10.1016/j.apsb.2022.02.002. Harrison, 2016, Nature Reviews Drug Discovery 15, 817 to 818, doi:10.1038/nrd.2016.184. DiMasi et al., 2016, Journal of Health Economics, summarized by the Tufts Center for the Study of Drug Development. Sena et al., 2007, as cited in a 2012 review in European Journal of Nuclear Medicine and Molecular Imaging, PMC3411287.
01
Slow
It takes about 10 years to bring one new medicine to patients (Tufts, 2016).
02
Expensive
Each approved medicine costs about $2.6 billion to develop, counting the candidates that fail (Tufts, 2016).
03
High failure rates
Nine in ten candidates fail somewhere between the first human dose and approval (Sun et al., 2022).
04
Poor early predictions
Of 494 stroke treatments that worked in animals, only 3 showed a convincing effect in patients (Sena et al., 2007).
Candidate screening
We computationally screen biologic candidates.
A ranking you can interpret
Testing every sequence in the lab is slow and expensive. With a ranking you can interpret, you test only the designs that rank highest, cutting lab expenditure, accelerating your research cycle, and reaching patent priority before your competitors.
Candidate B-07 Ranked 1 of 10,000
- Binding affinity
- 0.8 nM
- Developability
- 90/100
- Immunogenicity
- 0 epitopes
- Toxicity
- 0 of 24 off-targets
- Thermostability
- kcal/mol
- Target engagement
- 96 % occupancy
Hypothetical example for illustration only: invented data, not a real candidate. Folding is the variant’s folding free energy minus that of the starting antibody; negative values indicate stabilization.
Track record
Our technologies surpassed several world records , validated by an independent lab.
In the 2026 EGFR/cetuximab binder optimization competition, three of our ten designs surpassed the previous world best, and the candidate we ranked first proved to be the tightest binder measured, at 43.9 times tighter than the cetuximab starting antibody and 1.6 times tighter than the best prior design, all done in under one day.
The people
Meet the team
-
Founder and CEO
Founder and CEO/CTO; computational biology and bioinformatics at ETH Zürich with a background in math and computer science; leading the R&D and product development at Interlit.
-
Dr. Shen Yan
Co-founder and Chief Scientific Officer; PhD in Molecular and Translational Biomedicine at ETH Zürich; more than 15 years in the wet lab.
-
Felix
Senior associate; focuses on protein design.
-
Zuzi
Startup advisor with a background in psychology.
Supported by
From the blog
Our latest research notes.
- 30 September 2026Dissociation rate drives binding affinity differences among optimized cetuximab-derived EGFR binders
- 30 July 2026Two Scientific Innovators, One Shared Mission
- 18 July 2026The Score Most People Use to Rank Protein Binders Doesn’t Predict Binding
- 14 July 2026In Under a Day, We Set Three World Records in Protein Design
Questions
Frequently asked questions
Interlit is a Zürich company that develops novel methods, user-friendly software, and interpretable AI models to design biologic medicines.
We start from your target and your binding or screening data, then optimize and rank designs from there.
Each candidate comes with the assessments behind its rank: binding affinity, thermostability, immunogenicity, toxicity, developability, and target engagement. You can see why one design ranks above another.
We are taking early requests. Join the waiting list or ask about a screening project through the
Contact
Contact us or join our waiting list.
The values below guide our work and every product decision.
- Staying trueWe stay true to ourselves, to you and to our mission.
- CuriosityWe keep learning, exploring, and inventing new things, and we challenge assumptions.
- Thinking differentlyWe question established dogmas and the status quo.
- Quality and aestheticsWe believe great science deserves exceptional design, precision, and craftsmanship.
We are happy to hear from you. Fill out this short form to join the waiting list or send us a question.
Takes about 3 minutes
Increasing Humanity's Healthspan
Ready to turn early hit data into a life-saving candidate?
We specialize in computational and biophysical lead optimization. Input your target and starting structure alongside your binding or screening data, and we will model and prioritize your next-generation designs to test.