BEAM-AI

Snapshot Protein-Binders VHH/IgG

We use a structure-based, in-silico engine to predict, rank, and design high-probability binders and antigen pairs for you, then confirm them fast via outsourced assays. Capital-light. Cycle time short. Translational by design. We are looking for interested cooperation partners in academia and pharma/biotech.

Get in touch

Conceptual mockup. Features and visuals subject to change.

THE PROBLEM

Antibody design is slow and costly in addition to a high clinical attrition rate.

  • Traditional approaches have difficulties with certain antigens, and wet-lab-only paths burn years and budget with uncertain success.
  • New treatments such as multispecific ADCs and AND-gate CAR-T explode the search space, but traditional trial-and-error doesn’t scale.

With computational accuracy improving and validation becoming faster and cheaper, the field is primed for a shift from trial-and-error to compute-first discovery.

OUR SECRET SAUCE

BEAM-AI

BEAM-AI, short for Biomolecular Estimator of Atomic-group probabilities, is our structure-based engine for discovering and designing tumor-targeting binders.

It analyzes protein structures to predict where and how a molecule is most likely to bind, then generates binders and segments and ranks potential targets and binder sequences by binding probability and developability defining what to synthesize next.


This creates short, capital-light discovery cycles that translate computational predictions into validated biology within weeks, not years.

How it works

Input
3D protein structure, target preference location.

Output
Target Preference Map and a ranked set of binders or antigen pairs for downstream validation.

Validation
Results are confirmed through fast, outsourced biophysics and, upon request, cell based assays.

Meet our team

CEO

Stefan Munker

Stefan leads Oncera with a strong scientific vision, he is an MD and ensures that Oncera delivers commercial relevance and scientific impact.

Head of AI-Drug Discovery

Maximilian Kienlein

PhD, Physicist at LMU, is the technical lead with a clear vision to bring binder engineering to the customer with generative AI

Cofounder

Grzegorz Popowicz

PI, AI-Drug Discovery, Physicist (Co-Founder Khumbu.Ai) with a clear scientific vision to improve structure based drug discovery with deep learning approaches

Head of 3D Tumor-Avatarbank

Mari Hambardzumyan

MSc. Biotechnology, Lab Operations and Bioinformatics, Leads the wetlab validation of our AI model.

Advisor

Prof. Ekkehard Leberer

Founder of ELBIOCON Life Science Consulting, advising on nucleic-acid therapeutics, biologics, and drug delivery. Former senior R&D/external-innovation leader at Sanofi/Aventis (incl. Head of Biotechnology Germany); earlier Senior Research Officer at Canada’s NRC Biotechnology Research Institute.

Advisor

DR. sandeep kumar

Co-founder and CTO of a stealth-mode TechBio startup, while also consulting in protein design and AI/ML–driven drug discovery. Until April 2025, he was a distinguished Fellow at Moderna, where he led molecular modeling and biodesign innovation.

Advisor

Prof. Katja Hanack

Founder/CEO of new/era/mabs and Professor of Immunotechnology at the University of Potsdam (since 2015); antibody-discovery expert and co-inventor of high-throughput selection tech; consultant to Stanford’s SPARK program.

CONTACT

Get in touch

We collaborate with oncology biotechs, academic groups, and investors who believe binder generation should move faster and smarter.

You want to change current paradigms as to how drug discovery is being done and have experience in AI engineering or other skills, then do not hesitate to contact us!

You are a company and want to try our binders for your new ADC or CAR-T programs, want to validate targets with BEAM-AI, or are interested in joining our pre-seed round, lets connect.