Scala Biodesign

**Product and the concrete problem it solves.** Scala Biodesign attacks one of the most expensive bottlenecks in modern biotechnology: getting a protein to actually behave the way a drug developer or industrial manufacturer needs it to. A therapeutic antibody may bind its target beautifully in a computer but aggregate, misfold, or express at uselessly low yields in a bioreactor; an industrial enzyme may catalyze the right reaction but denature at process temperatures; a vaccine immunogen may present the wrong conformation. Historically, teams have fixed these problems by brute-force experimental screening — building and testing thousands of variants across many wet-lab iterations, a process that consumes months to years and enormous budgets, and that frequently fails. Scala's answer is **ScalaOS**, a computational protein-design platform that engineers and optimizes proteins *in silico* so that far fewer, far better candidates ever reach the bench. The pitch to a pharma or chemicals customer is concrete: stabilize a "difficult" or previously undruggable target, raise expression and manufacturability, improve binding or thermostability, and compress the design-build-test cycle — Scala says it often reaches an optimized design in a *single* cycle rather than many. In effect it sells design confidence: fewer experimental rounds, higher hit rates, and shorter timelines for turning a promising biologic into a developable, manufacturable product.

**Core technology and how it actually works.** ScalaOS is deliberately *not* a single black-box model. Scala's differentiation is a hybrid architecture that integrates three complementary sources of truth: (1) **physics-based atomistic modeling** — the biophysical energy calculations that Scala's scientific lineage helped pioneer, which reason about whether a designed sequence will actually fold and remain stable; (2) **evolutionary information** — the statistical signal embedded in millions of naturally occurring related proteins, which encodes what nature has found tolerable; and (3) **large language models and generative AI** tuned per modality, with distinct models for antibodies, enzymes, and vaccines rather than one generic protein model. By fusing physics, evolution, and machine learning, ScalaOS aims to avoid the characteristic failure modes of each approach used alone — generative models that hallucinate un-foldable sequences, or pure physics pipelines that are accurate but too slow and narrow. The platform is presented as embedding directly into pharma R&D workflows: a designer specifies objectives (stability, expression, affinity, immunogenicity) and ScalaOS returns optimized sequences prioritized for experimental success. The company frames the platform as the productization of more than a decade of protein-design research spanning dozens of laboratories and scores of scientific publications, translated from an academic method into industrial software.

**Market, customers, and go-to-market.** Scala sells into the biologics and industrial-biotech value chain — pharmaceutical and biotech companies developing antibody drugs, enzyme and gene-therapy developers, vaccine programs, diagnostics makers, and chemical corporations that use engineered enzymes in "green" bio-based manufacturing. Its go-to-market is a platform-and-partnership model rather than a pure drug-discovery play: ScalaOS is offered to partners who retain their own pipelines, positioning Scala as an enabling design layer across many programs simultaneously — an attractive structure because it can attach to numerous customers and molecules without Scala bearing full clinical risk. The most striking public go-to-market claim is enterprise penetration: Scala states that within roughly eight months of commercial availability, **nine of the world's top twenty pharmaceutical companies** had adopted ScalaOS, alongside leading chemical corporations. It has publicly highlighted a collaboration with **Boehringer Ingelheim**, describing work that stabilized difficult drug targets, improved protein expression levels, and accelerated discovery timelines. That combination — a horizontal platform that plugs into blue-chip pharma R&D, monetized across many partners — is the model AI-for-biology investors most want to see, because it de-risks revenue away from any single molecule's success.

**Traction, funding, and third-party validation.** Scala was founded in 2022 and exited stealth in September 2023 with a $5.5 million seed round to commercialize its protein-design approach. On March 31, 2026 it announced a **$16 million Series A led by Grove Ventures**, with participation from TLV Partners, Deep Insight, and the **Israel Innovation Authority**, bringing total funding to roughly **$21.5 million** since inception. The strongest third-party validation is less the dollar figure than the customer roster: onboarding a large share of the top-tier global pharma companies and a named Boehringer Ingelheim collaboration is meaningful commercial diligence for an early-stage platform, since large pharma does not adopt design tooling casually. The scientific pedigree adds further validation — the platform builds on a body of peer-reviewed protein-design work and academic methods that have been independently used and cited across the field. The principal calibration for a diligence reader is that headline traction claims (the "nine of top twenty" figure, revenue, and the depth versus breadth of each engagement) are company-stated and not independently audited; the quality of the Boehringer relationship and whether adoptions convert into recurring, expanding contracts are the facts that would confirm the narrative.

**Founders and team background.** Scala's team is its most defensible asset. It was co-founded by **Dr. Ravit Netzer (CEO)** and **Dr. Adi Goldenzweig (CTO)**, who developed the underlying computational protein-engineering methods as researchers in the laboratory of **Prof. Sarel Fleishman at the Weizmann Institute of Science**; Fleishman serves as the company's Chief Scientist. The Fleishman lab is one of the world's leading academic groups in computational protein design and biophysical modeling, and this lineage places Scala among a small set of companies commercializing genuinely frontier, physics-grounded design methods rather than repackaging open-source models. The founders' credibility is that they authored the science they are now productizing. The open questions typical of a Weizmann-style deep-tech spinout apply: the depth of the commercial, enterprise-sales, and biopharma-partnership bench beyond the scientific founders; current headcount (not publicly confirmed); and the organization's ability to scale delivery as enterprise demand grows.

**Competitive dynamics.** AI-driven protein design is one of the most crowded and fastest-moving frontiers in the life sciences, so Scala's moat rests on method integration and pharma traction rather than novelty alone. (1) Against fellow **Israeli protein-AI peers** — DenovAI Biotech (de novo antibody and binder design from first principles) and MNDL Bio (AI optimization of DNA constructs for expression) — Scala is differentiated by optimizing and re-engineering *existing* proteins and antibodies for developability across modalities, rather than pure de novo generation or DNA-level tuning. (2) Against well-funded **global generative-biology platforms** — Cradle (Netherlands), Generate:Biomedicines and Absci (US), and Nabla Bio — Scala's claimed edge is its physics-plus-evolution-plus-LLM hybrid and its embed-into-pharma-workflow positioning. (3) Against **foundation-model efforts** such as Isomorphic Labs and the AlphaFold/ESM structural-prediction lineage, Scala competes by focusing on actionable design and manufacturability rather than structure prediction alone. (4) It also competes with the **status quo** — customers' internal computational teams and traditional experimental directed-evolution screening. Its plausible edges are: (i) a world-class Weizmann scientific pedigree; (ii) a genuinely hybrid method rather than a single model; (iii) modality-specific models for antibodies, enzymes, and vaccines; and (iv) early blue-chip pharma adoption. The countervailing risk is that this is a field where deep-pocketed incumbents and foundation-model labs iterate rapidly and can compress differentiation.

**Defense, security, and resilience dual-use relevance.** Scala's dual-use relevance is real but should be read honestly as **biosecurity and medical-countermeasure adjacency rather than a fielded defense capability**. The core capability — rapidly engineering stable, manufacturable antibodies, vaccine immunogens, and enzymes with fewer experimental cycles — maps onto national bio-resilience and medical-countermeasure preparedness: the same tooling that stabilizes a therapeutic antibody could, in principle, accelerate design of countermeasures against emerging or engineered biological threats, improve the thermostability and field-manufacturability of vaccines and biologics for austere or contingency settings, and shorten response timelines in a biothreat or pandemic scenario. Thermostable, high-yield enzyme and protein design also supports resilient, distributed bio-manufacturing — a supply-chain-resilience theme relevant to sovereign biodefense capacity. The honest calibration: Scala is a commercial pharma-and-industrial platform with no disclosed defense contracts, biodefense programs, or government-security customers; its dual-use weight is a credible adjacency grounded in the biosecurity value of faster countermeasure design, not a demonstrated fielded capability. Its strategic relevance rises if it engages medical-countermeasure, pandemic-preparedness, or sovereign-biomanufacturing use cases explicitly.

**Growth stage, trajectory, and key diligence risks.** Scala reads as an **early-stage** deep-tech company with unusually strong signals for its stage: founded 2022, Series A in early 2026, roughly $21.5 million raised, a world-class scientific team, and enterprise traction that would be impressive even for a later-stage firm. The trajectory an investor would want to see is conversion of trial adoptions into expanding, recurring, multi-program contracts and independent evidence that ScalaOS materially improves partner outcomes. The key diligence risks are: first, **claim verification** — the "nine of top twenty" adoption figure, engagement depth, and revenue are company-stated and unaudited, and platform trials in pharma frequently do not convert to durable spend; second, **intense, well-capitalized competition** from global generative-biology platforms and foundation-model labs that can erode differentiation; third, **monetization model risk** — capturing durable value as an enabling design layer (versus owning drug assets) requires favorable contract structures and IP terms; fourth, **commercial-scaling risk** — translating a founder-scientist-led spinout into an enterprise-grade software-and-services organization; fifth, **team/opacity gaps** — headcount, unit economics, and the depth of named partnerships are not fully public; and sixth, on the strategic axis, the **dual-use thesis remains an adjacency** absent explicit biosecurity or medical-countermeasure engagements. Progression toward "mid" stage would be evidenced by disclosed multi-year enterprise contracts, quantified partner outcomes, a broadening revenue base, and any explicit move into biosecurity-relevant applications.