AI-powered therapeutic discovery

One platform.
Multiple therapeutic
opportunities.

SynFold is an AI platform that designs miniprotein therapeutics against any disease target. We start with cancer — and we have the results to prove the engine works.

How it works Get in touch →
83%
Of our designed proteins work in the lab
21d
From idea to a validated therapeutic candidate
60×
Cheaper to produce than conventional biologics
Therapeutic candidates — in testing
Candidate A
94
Candidate B
89
Candidate C
83
Candidate D
76
Candidate E
70
Effectiveness scoreAI-ranked
The problem

Millions of possibilities.
Few become medicines.

Most drugs that enter development never reach patients. The process is expensive, unpredictable, and takes over a decade. We are building a smarter way forward.

~90%
Of therapeutic candidates fail
Nine out of ten drugs that enter development never make it to patients. Most fail early, wasting years of work and billions of euros
$2.6B
To bring one medicine to market
The average cost of developing a new medicine has doubled in the last decade, making it almost impossible for smaller research teams to compete
10–15y
From lab to patient
For patients with aggressive cancers, waiting over a decade for a new treatment is not an option. We are compressing that timeline with AI
How it works

One engine. Any disease target.

SynFold's closed-loop AI platform designs, ranks, and validates miniprotein therapeutics against any extracellular target. Cancer is our entry point. The same engine runs for any disease where a target exists on the cell surface.

What we offer
Therapeutic discovery as a service
Universities, research institutes, and smaller biotech companies can use our platform to find therapeutic candidates for any target they are working on — without building the technology themselves.
Research · Biotech · Small Pharma
What we are building
Our own cancer drug
We are also developing our own therapeutic candidate for a type of breast cancer that currently has no targeted treatment. Once validated, we plan to license it to a major pharmaceutical company.
Roche · AstraZeneca · BMS
36+
Therapeutic candidates designed by AI
83%
Work when tested in the lab
7
Show anti-cancer activity
60×
Cheaper to produce than standard medicines
Our focus

Breast cancer's most aggressive
subtype has the fewest options.

Triple-negative breast cancer is defined by what it lacks — no hormone receptors, no approved targeted treatment, and survival rates that have barely improved in decades. Our lead therapeutic candidates show more than 50% cancer cell reduction in laboratory tests.

Step 01
Find the target
We identify a specific weak point on the surface of cancer cells — a protein the tumour depends on to grow and spread.
Target identification
Step 02
Design the drug
Our AI generates thousands of tiny proteins shaped to fit that weak point precisely — like designing a key for a very specific lock.
AI design
Step 03
Rank the best ones
The computer simulates how each candidate binds to the target and filters out the weakest options before any lab work begins.
Computational screening
Step 04
Test in the lab
The top candidates are produced and tested on cancer cells and 3D tumour models to confirm they work and are safe for healthy cells.
Lab validation
>50%
Reduction in cancer cells in lab tests
3 weeks
From computer design to lab-confirmed candidate
Late 2026
Patent filing planned
Pre-clinical
Stage before licensing to a pharma partner
Technology

The full stack,
built in-house.

Every step — from AI design to laboratory testing — is done by our team. No outsourcing. No black boxes. 83% of our designed proteins expressed successfully in bacteria. The machine works.

AI protein design
Structure prediction
Binding simulation
Molecular dynamics
3D tumour models
Cell viability assays
Gene expression analysis
Protein production
SynFold
Platform
Structure prediction
AI protein design
Sequence optimisation
Binding simulation
Molecular dynamics
3D tumour models
Cell testing
Work with us

The next generation
of cancer medicine
starts here.

We are looking for research partners, collaborators, and pharmaceutical companies who want to be part of what comes next in therapeutic discovery.

Catarina Tavares Cavaleiro
SynFold Therapeutics
ITQB NOVA · Universidade Nova de Lisboa · GIMM
linkedin.com/in/catarinatavarescavaleiro →
Connect on LinkedIn How it works →
MIT €100K Finalist 2026 · Health Innovation Award 2025 · Lisbon, Portugal