deepmirror
deepmirror is a London-based AI drug discovery SaaS platform that combines generative molecular AI, ADMET/potency prediction, and structure-based co-folding design to accelerate hit-to-lead optimization for biotech, pharma, academic, and global health researchers.
- Company typePrivate
- Founded2019
- HeadquartersLondon, United Kingdom
- Headcount1–10
- GTM typeB2B
- OfferingSoftware
What deepmirror does
deepmirror is a London-based AI drug discovery software company founded in 2019 that builds a cloud-based platform combining generative molecular AI, predictive modeling for potency and ADMET properties, and structure-based co-folding design. The unified platform targets medicinal chemists and drug discovery teams at biotech and pharmaceutical companies, enabling hit-to-lead and lead optimization workflows — including billion-molecule virtual screens that the company has demonstrated completing in hours with 60%+ hit rates on challenging transcription factor targets, and fine-tuned predictive models validated to triple R² accuracy on customer-specific datasets. The company is ISO 27001 certified and integrates third-party foundation models such as Boltz-2 (MIT/Recursion) for combined co-folding and affinity prediction at roughly 1000x the speed of traditional physics-based methods.
Revenue is generated through a hybrid SaaS model: quote-based commercial subscriptions (monthly or annual billing) for biotech and pharma customers, plus usage-based API access. A free tier serves academia and, through the March 2026 launch of the dd4gh (Drug Design for Global Health) platform co-developed with Medicines for Malaria Venture and funded by the Gates Foundation, non-commercial researchers in low- and middle-income countries working on malaria, tuberculosis, and neglected tropical diseases. Named commercial customers include Tes Pharma, Curie.Bio, Cerevance, and Morphic, alongside the University of Manchester and Sanders Tri-Institutional Therapeutics Discovery Institute.
The company is venture-backed with approximately $2.4M in pre-seed funding (2024) plus a several-million-USD 2025 strategic round from Goertek, BYD, and a Hong Kong family office directed at a separate spatial and embodied intelligence initiative. It is led by three co-founders — Max Jakobs (CEO), Andrea Dimitracopoulos (COO), and Ryan Greenhalgh (CTO) — and supported by an unusually credentialed advisory board including alumni of Relay Therapeutics, BMS, GSK, Exscientia, and Atavistik Bio. The platform is accessible via a self-serve web app and a Python SDK/API, with integrations into CDD Vault and Google Cloud infrastructure.
deepmirror firmographics
Firmographics- Name
- deepmirror
- Legal name
- deepmirror Ltd
- Website
- https://deepmirror.ai
- Company type
- Private
- Founded year
- 2019
- Operating status
- Operating
- Headcount range
- 1–10 employees
- Short description
- deepmirror is a London-based AI drug discovery SaaS platform that combines generative molecular AI, ADMET/potency prediction, and structure-based co-folding design to accelerate hit-to-lead optimization for biotech, pharma, academic, and global health researchers.
- Ownership category
- akta.pro rank
deepmirror industry classification
Industry- Product category
- AI drug discovery software
- NAICS
- Software Publishers (5132), Computer Systems Design and Related Services (54151)
- SIC
- Laboratory Analytical Instruments (3826)
- akta.pro primary industry
- AI/ML Platforms for Drug Discovery & Experiment Optimization (HLAGAJAO)
- akta.pro secondary industry
- Cheminformatics & Molecular Modeling / CADD (HLAGAJAE)
Keywords
Where deepmirror is headquartered
LocationHeadquarters
- HQ city
- London
- HQ country
- United Kingdom
- HQ region
- Europe
Offices1 record
Markets served
deepmirror business model
Business model- GTM type
- B2B
- Offering type
- Software
- Cost components
- Personnel, Technology or R&D, Infrastructure, Marketing or Sales
Revenue model
- SaaS Platform Subscription (Commercial): deepmirror operates as a cloud-based SaaS platform offering subscription-based access to its AI drug discovery tools. Commercial customers (biotech, pharma) access the platform on a subscription or usage basis with tiered plans. Terms reference 'Fees' due in advance with automatic billing.
- Free Access for Global Health / Academia: The dd4gh platform (Drug Design for Global Health) is provided entirely free of charge to eligible researchers working on malaria, tuberculosis, and neglected tropical diseases. Funded in part by the Gates Foundation. Academic users can also access the platform at no cost.
- API / Usage-based Access: Platform access via API enables programmatic screening of large molecular libraries (e.g., 1 billion molecules), with infrastructure automatically scaling to process hundreds of millions of molecules in parallel. Usage is metered through the deepmirror API.
Pricing tiers
| Model | Billing | Price |
|---|---|---|
| Freemium | Annual | Free tier for global health researchers and academia |
| Subscription | Monthly | Commercial SaaS platform subscription |
| Subscription | Annual | Enterprise tier with full platform access |
Go-to-market motion2 records
Distribution channels4 records
Marketing channels6 records
deepmirror product offering
Product offeringCore offering
deepmirror is a cloud-based AI drug design platform that combines generative AI, data-driven predictive models for potency and ADMET properties, and structure-based co-folding algorithms to help medicinal chemists prioritize and design promising drug candidates. It enables billion-scale virtual screening, uncertainty-aware hit prioritization, and structure-guided lead optimization for biotech, pharmaceutical, academic, and global health drug discovery programs.
Product overview
deepmirror offers an AI-powered drug design platform with three core integrated capabilities: (1) the deepmirror App as the primary interface, (2) the dd4gh platform for global health drug discovery, and (3) supplementary tools including Generative AI Engine, Data-Driven Models, Structure-Guided Design, Boltz-2, and API/SDK access. The platform accelerates hit-to-candidate optimization by combining predictive analytics for potency and ADMET properties, generative AI for novel molecule creation, and structure-based design tools. The dd4gh platform extends these capabilities to non-commercial researchers in low- and middle-income countries working on malaria, tuberculosis, and neglected tropical diseases.
Differentiator
Problem solved
Functional benefit
Brands
- dd4gh (Drug Design for Global Health): Free AI drug discovery platform for non-commercial researchers working on malaria, tuberculosis, and neglected tropical diseases, developed in partnership with Medicines for Malaria Venture.
Products and services
- deepmirror (DeepMirror App) AI-powered SaaS platform that enables medicinal chemists to predict molecular properties (potency, ADMET), generate novel drug candidates, and screen billion-scale molecular libraries. Used by biotech and pharma teams for hit-to-lead and lead optimization.
- Drug Design for Global Health (dd4gh) Free AI-powered drug discovery platform developed in partnership with Medicines for Malaria Venture (MMV) for non-commercial researchers in low- and middle-income countries working on malaria, tuberculosis, and neglected tropical diseases. Combines predictive and generative AI with active learning.
- deepmirror API Programmatic API for automated screening of ultra-large molecular libraries, predictive modeling, and batch inference. Enables integration of deepmirror's AI capabilities into customer drug discovery workflows with automatic infrastructure scaling.
- deepmirror-client SDK Python SDK providing programmatic access to the deepmirror platform. Enables developers to build predictive models, run batch inference, screen molecular libraries, and access protein-ligand co-folding functionality.
Quantifiable outcome
- 60%+ hit rate from billion-scale virtual screen (vs. typical 5-10% for transcription factor targets)
- +6 more outcomes
Companies that use deepmirror
Customer profileNamed customers9 records
Segments4 records
Ideal customer profiles3 records
deepmirror technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- Yes
- API docs
- API detail
Core technology
AI maturity
App detail
Integration1 record
AI capability6 records
Feature7 records
deepmirror partnerships and signals
Strategic signalPartnerships
Five partnerships are on record, tiered flagship / core, core, key technology, infrastructure and product co-development.
- Medicines for Malaria Venture (MMV)flagship / coredeepmirror and MMV co-developed Drug Design for Global Health (dd4gh), a free AI-powered platform for malaria, tuberculosis, and neglected tropical disease research. The partnership includes co-creation workshops in Ghana and Switzerland, with funding from the Gates Foundation. The platform launched in March 2026 and is available free to eligible researchers in LMICs.
- Collaborative Drug Discovery (CDD)coreStrategic partnership integrating deepmirror with CDD Vault, enabling researchers to connect cutting-edge AI to drug discovery data. CDD Vault manages chemical registration, SAR, and bioactivity data for 700+ research centers with 4B+ validated data points and 80M+ proprietary unique structures.
- MIT / Recursion (Boltz-2)key technologyBoltz-2, developed by MIT and Recursion Pharmaceuticals, is integrated into the deepmirror platform. It is the first public model combining co-folding with binding affinity predictions, achieving speeds approximately 1000x faster than traditional physics-based methods.
- Google CloudinfrastructureCollaboration with Google Cloud enables Curie.Bio to run advanced protein structure models (AlphaFold3, Boltz-2, Chai-1) on challenging, undrugged targets. Google Cloud provides scalable computational infrastructure for deepmirror's cloud-based operations.
- Sanders Tri-Institutional Therapeutics Discovery Institute (TDI)product co-developmentCollaboration with TDI (affiliated with Weill Cornell Medicine, Rockefeller University, Memorial Sloan Kettering Cancer Center) to fine-tune deepmirror's solubility predictions. Fine-tuning increased R² from 0.2 to 0.6, demonstrating improved predictive performance.
Scale indicators7 records
Recent moves6 records
Expansion highlights6 records
deepmirror competitors and assessment
Company assessmentDirect peers
- XtalPi: AI + quantum physics + robotics drug discovery platform serving pharma and biotech. Comparable in offering predictive modeling and generative molecule design alongside large-scale computational infrastructure.
- Atomwise: Pioneer in AI small-molecule discovery using convolutional neural networks for virtual screening. Directly comparable to deepmirror's generative and predictive chemistry capabilities, with multi-year pharma partnerships and broader target-class coverage.
- Insilico Medicine: AI-driven drug discovery platform combining generative chemistry, predictive modeling, and target identification. Like deepmirror, it serves biotech and pharma customers seeking to compress hit-to-lead timelines, but operates at significantly larger scale with internal pipeline assets.
- Iambic Therapeutics: AI-driven small-molecule discovery combining foundation models for chemistry with physics-based simulation. Direct competitor in generative + predictive drug design for biotech customers, with overlapping scientific philosophy around uncertainty-aware predictions.
- Genesis Therapeutics: AI-native drug discovery company combining deep learning for protein structure prediction with generative chemistry. Direct overlap with deepmirror's cofolding and structure-guided design capabilities.
- Exscientia: AI-first drug discovery company combining generative AI, active learning, and automated synthesis. Deepmirror's advisor Jérémy Besnard is an Exscientia co-founder, signaling deep overlap in platform philosophy and customer base of biotech and pharma partners.
Broad incumbents
- Recursion Pharmaceuticals: AI-driven biotech combining high-throughput biology, generative chemistry, and co-developed models (e.g., Boltz-2 with MIT). Overlaps with deepmirror in generative chemistry and structure-based design while operating a much broader internal pipeline.
- Schrödinger: Established physics-based and AI-augmented molecular simulation platform serving most large pharma companies. Competes in the same computational chemistry and SBDD workflow that deepmirror targets, but with a much broader portfolio (materials, biologics) and deeper enterprise install base.
Regional players
- Cresset: Computational chemistry and molecular design software provider with strong UK/European footprint. Comparable customer base of medicinal chemists in biotech and pharma, though traditionally more ligand-based/field-based than generative AI focused.
Emerging players
- Cradle Bio: AI protein engineering platform using generative models for biologics design. Adjacent rather than identical, but competes for the same AI-for-drug-discovery customer mindshare and uses similar foundation-model + fine-tuning architecture as deepmirror.
Market position
Strengths5 records
Weaknesses5 records
Competitive moat6 records
Key risks6 records
Key highlights7 records
Customer concentration
deepmirror social profiles
Digital presencedeepmirror compliance and trust
Trust signalCompliance1 record
deepmirror financial estimates
Financial estimateRevenue estimate
Valuation estimate
deepmirror leadership team
Management profileNumber of profiles
Profiles4 records
deepmirror funding detail
Funding detailFunding overview
Funding rounds4 records
Investors2 records
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
deepmirror M&A and investment
M&A and investmentM&A
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about deepmirror
What does deepmirror do?
deepmirror is a cloud-based AI drug design platform that combines generative AI, data-driven predictive models for potency and ADMET properties, and structure-based co-folding algorithms to help medicinal chemists prioritize and design promising drug candidates. It enables billion-scale virtual screening, uncertainty-aware hit prioritization, and structure-guided lead optimization for biotech, pharmaceutical, academic, and global health drug discovery programs.
Is deepmirror a public or private company?
deepmirror is a private company. It is classified as venture growth investor backed and is currently operating.
When was deepmirror founded?
deepmirror was founded in 2019. It employs 1 to 10 people.
Where is deepmirror based?
deepmirror is headquartered in London, United Kingdom, in the Europe region.
How does deepmirror make money?
Three revenue lines are on record. SaaS Platform Subscription (Commercial) is the primary driver. The others are free Access for Global Health / Academia and API / Usage-based Access.
Who are deepmirror's main competitors?
Direct peers on record are XtalPi, Atomwise, Insilico Medicine, Iambic Therapeutics, Genesis Therapeutics and Exscientia. Broad incumbents are Recursion Pharmaceuticals and Schrödinger. Cresset is listed as a regional player. Cradle Bio is listed as an emerging player.
Does deepmirror have an API?
Yes. deepmirror provides a public API that enables programmatic access to its drug discovery platform. The API allows users to screen ultra-large molecular libraries (billions of compounds), build predictive models, and generate molecule predictions. Developers can use the deepmirror-client Python SDK to access these capabilities. Example use cases include batch inference for screening billion-scale libraries and running Boltz-2 for structure prediction. Developer documentation is at github.com/deepmirror/deepmirror-client.
What industry is deepmirror in?
deepmirror's product category is AI drug discovery software. Its primary akta.pro industry code is HLAGAJAO, AI/ML Platforms for Drug Discovery & Experiment Optimization, with a secondary code of HLAGAJAE, Cheminformatics & Molecular Modeling / CADD. Its NAICS code is 5132 and its SIC code is 3826.