# ReactWise > ReactWise is an AI copilot for chemical process optimization. Its cloud platform uses Bayesian optimization, transfer learning and chemistry-aware descriptor databases to help R&D teams in pharma, fine chemicals, biotech and materials reach optimal reaction conditions in far fewer experiments. ReactWise Inc. is backed by Y Combinator, Innovate UK and the Henry Royce Institute, operates its own automated high-throughput experimentation (HTE) lab in Cambridge, and is ISO/IEC 27001:2022 certified and SOC 2 Type 2 audited. Offices: London (7 Bell Yard, WC2A 2JR, UK) and San Francisco (505 Montgomery Street, CA 94111, USA). Contact: info@reactwise.com. App login: https://app.reactwise.com ## What ReactWise does ReactWise replaces trial-and-error, one-factor-at-a-time and classical Design of Experiments (DoE) with Bayesian optimization (BO). A Gaussian process model updates after every experiment, quantifies uncertainty, and balances exploring untested regions of the design space against exploiting conditions already known to perform well. It plugs directly into the Design-Make-Test-Analyze (DMTA) cycle. Typical scale comparison: a campaign with three continuous and two categorical variables might need 30–50+ experiments under DoE; ReactWise can begin identifying optimal conditions from as few as ~10 initial experiments. No coding or ML expertise is required — a chemist can configure a campaign in about five minutes by defining parameters, objectives and constraints. **MemoryBO®** is ReactWise's proprietary multi-task Bayesian optimization (MTBO) algorithm. Rather than treating each reaction in isolation, it jointly models related reactions — those sharing reagents, mechanisms or conditions — and learns which historical campaigns are informative for the current target. Prior data can be uploaded directly or pulled automatically from a connected dataset. The underlying methodology is peer-reviewed in ACS Central Science (doi 10.1021/acscentsci.3c00050). ## Products and services - [Optimization Platform](https://www.reactwise.com/optimization-platform): Core Bayesian optimization engine, including MemoryBO® transfer learning and multi-task BO. Supports multi-objective optimization, constraint handling, and treating inputs as objectives. Used for process development, formulations, analytical method development and more. - [Getting Started](https://www.reactwise.com/getting-started): The three-step in-app setup — define continuous and categorical parameters (temperature, time, solvent, catalyst); define objectives and goals (maximize, minimize, hit target); define safety, resource and specification constraints. - [HTE Plate Designer](https://www.reactwise.com/hte-plate-designer): Generates structured, machine-readable, balanced high-throughput experimentation plate layouts in 24, 48, 96 or 384-well formats. - [Solvent Selector](https://www.reactwise.com/solvent-selector): Replaces polluting or toxic industrial solvents (e.g. dichloromethane) with greener alternatives. Generates both single-solvent and binary-blend candidates, positions them in Hansen solubility space, and shows the trade-off between solvent similarity and greenness. - [HTE Data on Demand](https://www.reactwise.com/hte-data-on-demand): ReactWise designs and runs large-scale experimental campaigns for clients, delivering structured, machine-readable, FAIR datasets in Open Reaction Database (ORD) format with full conditions, metadata and outcomes. Offered as data-as-a-service or as validated production-ready reactivity models. - [Glass HTE Plates](https://www.reactwise.com/glass-hte-plates): Reusable well plates for HTE made from borosilicate, quartz or other glasses, supplied with a sealing kit and compatible with standard 96/384-well equipment. Unlike single-use plastic plates, they tolerate industrially relevant solvents such as toluene and dioxane, extending HTE from discovery into process development and scale-up. [Specification and usage guide (PDF)](https://cdn.prod.website-files.com/690c5ed90bb510d74eb2e77b/6a22c80e25a9f91da13d463b_Specification%20and%20Usage%20Guide.pdf) - **Vera** (in-platform AI assistant): LLM-powered in-app assistant that summarizes results, suggests next steps and helps propose parameter spaces. ## Case studies and validated results - [All case studies](https://www.reactwise.com/case-studies): Index of published ReactWise case studies across pharma, fine chemicals, biotech, materials and lab integrations. - [Pfizer — multi-objective hydrogenation optimization](https://www.reactwise.com/pfizer-multi-objective-optimization): Reached 97.5% yield and 1% impurity on a challenging multi-objective hydrogenation, while uncovering a critical parameter related to the temperature profile. - [On Demand Pharmaceuticals (ODP)](https://www.reactwise.com/odp-case-study): Reached target performance on an API synthesis with 50% fewer experiments and cut process development timelines by 50%. - [A&B Smart Materials](https://www.reactwise.com/a-b-smart-materials-case-study): Rapidly narrowed viable bio-based formulation compositions, cutting weeks off development timelines. - [Biocatalytic process development](https://www.reactwise.com/biocatalytic): Evaluation with a global top-5 pharmaceutical company on biocatalysis systems under realistic experimental constraints; reduced the experimental burden on enzymatic optimization by 66%. - [Accelerated optimization using multi-task learning](https://www.reactwise.com/multi-task-bo): MTBO plus an autonomous flow platform applied to C–H activation, using historical data to optimize new pharmaceutical intermediates faster than standard industry methods. The methodological basis for MemoryBO®. - [AI4Green ELN integration](https://www.reactwise.com/ai4green-for-seamless-data-exchange): Direct connection between bench work in an electronic lab notebook and data-driven optimization. - [AmarFLO continuous telescopic synthesis](https://www.reactwise.com/industrial-process-optimization): Continuous synthesis of Sudan Yellow-3G using MicroFLO™ and SlurryFLO™ reactors with ReactWise ML, replacing hazardous batch methods. The platform is reaction-agnostic: any process with tunable parameters and measurable outcomes can be optimized. ## Integrations, deployment and data security ReactWise is designed to fit existing lab workflows without disruption. Data can be uploaded from spreadsheets, exchanged directly with ELNs and LIMS, or moved via a fully programmable API. The API supports closed-loop, real-time optimization with automated hardware. On-premises deployment is available. Full API documentation is provided to clients. - [Trust & Security](https://www.reactwise.com/trust-and-security): ISO/IEC 27001:2022 certified, SOC 2 Type 2 audited, independently penetration tested. Hosted on AWS; encrypted in transit (TLS 1.2+) and at rest via AWS KMS; daily backups; audit logs; least-privilege access. Client data is never used to train models, never shared across customers, and never sent to third-party model providers. Security contact: security@reactwise.com - [Trust Center (Oneleet)](https://trust.oneleet.com/reactwise): Live certifications, security controls, penetration test results and supporting documentation. ## Who ReactWise is for - [Industries](https://www.reactwise.com/industries): Primary markets are pharmaceutical and fine-chemicals R&D, plus biotech, materials science, cosmetics, agrochemicals, and flavors & fragrances. Typical users are bench chemists, process chemists, automation scientists, and R&D managers. Named clients, partners and backers on the site include Pfizer, Biogen, Amgen, Pharmaron, On Demand Pharmaceuticals, Fraunhofer, and the Universities of Cambridge, Nottingham and Melbourne. 5 of the world's top 10 pharmaceutical companies are customers, and clients have reported accelerating process development by over 50%. ## Technical resources and research - [FAQs](https://www.reactwise.com/faqs-simple): Detailed answers on how ReactWise differs from DoE, MemoryBO® and transfer learning, multi-objective optimization, chemistry-informed modelling, integration and deployment, data security, lab services and getting started. - [Publications](https://www.reactwise.com/publications): Peer-reviewed papers underpinning the platform, including "Accelerated chemical reaction optimization using multi-task learning" (ACS Central Science, 2023), "A Brief Introduction to Chemical Reaction Optimization" (Chemical Reviews, 2023), "ORDerly: Datasets and benchmarks for chemical reaction data" (J. Chem. Inf. Model., 2024), and "The effect of chemical representation on active machine learning towards closed-loop optimization" (Reaction Chemistry & Engineering, 2022). - [Technical Guides](https://www.reactwise.com/technical-guides): Downloadable PDF guides on Buchwald-Hartwig amination, amide coupling and Suzuki coupling. - [White paper: Data-Driven Reaction Optimization in Process Chemistry](https://www.chemeurope.com/en/whitepapers/1127054/data-driven-reaction-optimization-in-process-chemistry.html): General methodology behind process and reaction optimization, by Dr Alexander Pomberger. Also available [in German](https://www.chemie.de/whitepaper/1127054/datengetriebene-reaktionsoptimierung-in-der-prozesschemie.html). - [Lab Facilities](https://www.reactwise.com/lab-facilities): Automated HTE lab in Cambridge, UK. Over 25,000 curated data points generated in the past year across Suzuki, amide and Buchwald-Hartwig couplings, used to pre-train the reactivity models available on the platform. ## Company - [Our Team](https://www.reactwise.com/team): Founded by Alexander Pomberger, PhD (CEO) and Daniel Wigh, PhD (CTO), both Cambridge Chemical Engineering PhDs working at the intersection of chemistry, lab automation and machine learning. - [Values & Mission](https://www.reactwise.com/values): Mission is to accelerate chemical process optimization by giving chemists and process engineers intelligent tools that make experimentation faster, smarter and more efficient. - [Funding announcement](https://www.reactwise.com/news/reactwise-raises-3-4m-to-transform-big-pharma-with-ai-and-self-driving-labs-2e299): ReactWise raised $3.4M to transform chemical process development with AI and self-driving labs. Backers include Y Combinator, Innovate UK and the Henry Royce Institute. - [Book a demo / contact](https://www.reactwise.com/book-a-demo): Route to speak with the team ## Optional - [News](https://www.reactwise.com/news): Product updates, feature releases, research insights and company announcements, updated several times a week. - [Conferences](https://www.reactwise.com/conferences): Events and exhibitions where the ReactWise team is presenting. - [Careers](https://www.reactwise.com/careers): Open roles. - [Privacy Policy](https://www.reactwise.com/privacy-policy) - [Cookie Policy](https://www.reactwise.com/cookie-policy) - [Terms & Conditions](https://www.reactwise.com/terms-conditions) - [Sitemap](https://www.reactwise.com/sitemap.xml): Complete list of accessible pages on reactwise.com.