Last updated September 24, 2026

Orbion

Using AI to predict protein expression, stability and developability before lab work

Protein DesignDrug DiscoveryLife Science Tools

Protein structure prediction is now widely accessible, but choosing which will express well and remain stable still requires repeated lab work. For difficult targets, poor construct choices can delay experiments and consume weeks of production effort before a usable sample is ready.

Orbion predicts expression, stability and from amino-acid sequence so teams can rank constructs and mutations before lab work. Its subscription software helps biotech, pharma and research teams shorten protein-development cycles and reduce experimental iteration.

Key Investor Signals

Co-founder scaled Twentify to 100+ customers and 210,000+ users

Çağlar Bozkurt co-founded in 2014; by 2016 it had 100+ enterprise customers and 210,000+ app users, showing prior software scale-up experience.

Research collaboration with leadXpro on membrane proteins

Orbion and Swiss structural-biology specialist leadXpro are testing Astra AI models to improve work on difficult membrane proteins, combining Orbion’s ML with leadXpro’s experimental platform.

Prospective FGF21 test doubled usable concentration

In a prospective n=1 comparison, Orbion’s design concentrated to 1.4 mg/mL versus 0.7 mg/mL for the literature construct, despite producing 2.4× less purified protein.

Other Positive Signals

  • Benchmarked Astra models across 18,367 proteins in five 2026 protein-family studies.
  • Raised €100,000 from in January 2025
  • certified with EU-hosted customer data processing.
  • Received Nucleate Germany’s 2026 Scientific Excellence Award.
  • invested in Orbion through its Swiss accelerator in January 2026.
  • Selected for QAI Ventures' 2026 Swiss quantum and AI accelerator cohort.

Team

Aniruddh Goteti

Co-Founder and Co-CEO

  • Built production machine-learning systems at Blue Yonder, applying AI to enterprise supply-chain software.
  • Holds an MSc in Computer Science from Blekinge Institute of Technology and a 2024 MBA from WHU.
  • Chaired the WHU Entrepreneurship Roundtable, leading a seven-person MBA team and building its founder community.

Çağlar Bozkurt

Co-Founder

  • Co-founded Twentify, which by 2016 had reached 100+ enterprise customers and 210,000+ app users.
  • Led product growth at Taxfix, driving 30% more referral purchases and over 20% higher visit-to-registration conversion.
  • Served as Senior Product Manager at AMBOSS, bringing direct experience building software for healthcare professionals.

Product

Orbion scores protein sequences before lab work, predicting expression, stability and developability so teams can choose what to make and which mutations to test. It also generates literature-grounded experimental protocols for selected constructs.

Protein construct ranking

Scores candidate sequences for predicted expression, solubility and aggregation risk, helping teams prioritize the strongest constructs before committing time and resources to laboratory production.

Mutation stability prediction

Ranks point mutations by predicted impact on thermal stability and folding, helping teams identify substitutions most likely to improve protein stability before experimental screening.

Lab protocol generation

Builds literature-grounded protocols covering vector choice, expression host and key production steps, giving researchers a practical starting plan for selected protein constructs.

Market Overview

often requires repeated rounds of design, expression and testing before a stable, usable construct is found. The bottleneck is especially acute for membrane proteins and other hard-to-express targets, where failed experimental cycles consume time and specialist lab capacity.

Commercial adoption is emerging among biotech and pharma teams. reported 21+ customers, including Novo Nordisk and Johnson & Johnson Innovative Medicine, before raising a $73M Series B in 2024. It also reported 1.2–12× faster R&D and cost reductions of up to 90% across customer projects.

Biologics' share of US drug spending
51%

accounted for 51% of U.S. drug spending in 2024, despite only 5% of prescriptions.

Cradle-reported R&D cost reduction
Up to 90%

Cradle reports AI protein engineering cut project costs by up to 90% across customer programs.

Protein engineering market by 2030
$7.6B

Grand View Research projects the global protein-engineering market to reach $7.6B by 2030.

Cradle's AI protein engineering Series B
$73M

Cradle raised $73M in 2024 after growing to 21+ customers using its AI protein-engineering platform.

Comparable Outcomes

AI protein design

Absci

IPO at ~$1.45B valuation.

built an integrated drug-creation platform combining synthetic biology and AI to discover and engineer protein therapeutics. Absci listed on Nasdaq in July 2021 at $16 per share, implying an IPO valuation of about $1.45B.

Protein engineering

Codexis

IPO at ~$443M valuation.

developed engineered enzymes and its CodeEvolver platform for biocatalysis, including pharmaceutical manufacturing. Codexis listed on Nasdaq in April 2010 at $13 per share, implying an IPO valuation of about $443M.

Interested in More Startups Like Orbion?

Get a curated pipeline of healthcare and life sciences startups aligned with your investment thesis.

Get Deal Flow

Sources

This profile was built from public company materials, ecosystem sources, market references, and professional profiles.

Get the DealBio Pulse

Weekly insights on health and longevity startups, standout founders, key markets, and notable exits.

You can unsubscribe at any time.