Monolith

What Monolith does

Monolith is an industrial AI company that provides purpose-built AI solutions and a cloud platform for engineers. It helps engineers transform complex engineering, test, and operational data into models that enable faster learning, smarter decisions, and fewer physical iterations. The company offers intuitive AI tools with a notebook interface, unique AI algorithms specifically for engineering applications, and an enterprise SaaS platform for large data, high-performance computing, and collaboration. Monolith also provides expert AI consulting and implementation services to guide AI adoption, instill best practices, and ensure success. Their solutions are designed to accelerate product design and validation, optimize test plans, find root causes of failures, and improve product designs.

Products

• Monolith AI PlatformA scalable cloud platform for large data, high-performance computing, and collaboration, designed for engineers to build self-learning models from test data.
• Test Plan Optimisation ModuleAI-guided tool to create more efficient battery test plans, reduce test steps by up to 70%, and increase test coverage using proprietary recommender algorithms.
• Test Data Validation ModuleAI-guided anomaly detector that pinpoints test errors in seconds, finding more than 90% of known issues across hundreds of signals to eliminate wasted test runs.
• System Calibration ModuleAI-guided optimization tool to calibrate complex systems faster for key performance requirements, balancing hundreds of input conditions and parameters to find optimal values.
• Next Test RecommenderAn AI-driven feature within the Test Plan Optimisation Module that guides testing strategy by applying thousands of algorithms to explore design space and recommend efficient test conditions.
• Anomaly DetectorA proprietary algorithm refined with customer test applications and data to find multiple error types across hundreds of channels in seconds, validated to find over 90% of known errors.
• Parallel CoordinatesA unique visualization tool developed for optimization, allowing users to specify target values with options for achieving them.

Key capabilities

• Intuitive AI tool with notebook interface
• Unique AI algorithms built for engineering applications
• Enterprise SaaS platform for scalability and collaboration
• Self-learning models trained from test data
• Optimise test plans
• Automate data inspection with AI anomaly detector
• Find root causes faster
• Calibrate complex systems
• Accelerate AI adoption with interactive notebooks
• Train and evaluate self-learning models without coding
• Understand product designs and parameter influence
• Predict performance for any test condition
• Dedicated spaces for teams to organize data, notebooks, and models
• Share insights through interactive dashboards
• Industry-standard security measures, data encryption, ISO 27001, and access control
• Multiple configurations and install options
• In-product tutorials, online learning tools, and custom training options
• Proprietary tool applies thousands of recommender algorithms
• AI-guided anomaly detector finds over 90% of known issues
• Interactive data visualiser for anomalies
• Automate detection process
• AI-guided optimisation tools for system calibration
• Build reliable virtual sensors from existing test data
• Cut total test time by up to 20x with targeted cycle design
• Populate calibration maps and lookup tables faster with machine learning
• Variational autoencoders for generative design and geometry parametrization
• Explainable AI functions

Use cases

• Accelerating product design and validation
• Optimizing test plans and reducing physical testing
• Finding hidden errors in data
• Streamlining test plans
• Finding root causes of failures
• Improving product designs
• Analysing battery cells undergoing aging tests
• Optimizing B-sample battery aging tests
• Identifying faulty assumptions, unnecessary test conditions, and data errors
• Simulating and validating vehicle performance
• Optimizing gas meter behavior and accuracy
• Packaging optimization
• Optimizing particle size and shape distribution for target dissolution profiles
• Developing next generation smart meters
• Cell-level fault isolation in battery pack validation
• Improving vehicle development by reducing physical tests
• Battery anomaly detection
• Predicting sloshing noise and reducing testing in automotive
• Accelerating pharmaceutical testing
• Cutting car setup time in motorsports
• Generative design of turbomachinery blades
• Improving surrogate model accuracy
• Integrating machine learning into CFD simulations
• Reducing fuel consumption by testing engines with AI
• Predicting dissolution kinetics of corticosteroid particles in nasal sprays
• Aircraft wind tunnel testing
• Flight dynamics prediction
• EV battery testing and validation
• Engine calibration

AI approach

Monolith provides an AI software platform and consulting services specifically for engineering applications. They build purpose-built AI solutions with unique algorithms for engineering, enabling engineers to train machine-learning models to optimize test plans, automate data inspection, find root causes, and calibrate complex systems. They emphasize user-friendly AI tools for domain experts, reducing the need for coding or data science PhDs. Their advanced research team also collaborates with customers to develop bespoke AI algorithms and solutions.

Tech named: deep learning, machine learning, AI algorithms, self-learning models, anomaly detection algorithm, recommender algorithms, random forest regression model, Bayesian Deep Learning, autoencoders, variational autoencoders, Neural Network, Gaussian Process Regression, Explainable AI

Industries served

• Software Development
• Aerospace Engineering
• Automotive Engineering
• Mechanical Engineering
• Industrial
• Motorsport
• Packaging
• Pharmaceutical
• Defense
• Energy

What it says sets it apart

• Purpose-built AI solutions for engineering workflows
• Intuitive AI tool with a notebook interface designed for domain experts
• Unique AI algorithms specifically for engineering applications
• Enterprise SaaS platform for large data, high-performance computing, and collaboration
• Expert AI consulting and implementation services (Success360 program)
• Refined from hundreds of AI projects with industry-leading engineering teams
• Proprietary algorithms tuned for real-world engineering test data challenges
• Ability to train machine-learning models without arduous coding or data science PhD required
• Focus on reducing physical iterations, development time, and costs
• Academic background with spin-off from PhD project on uncertainty quantification
• Innovate UK grant for Explainable AI research
• Partnerships with universities (Imperial, Cambridge, ISAE-Supaéro) for AI capability development
• Proprietary algorithms handling 3D data (patent pending)
• Members and contributors of NAFEMS
• Co-creation approach with customers to evolve the platform and address real-world challenges
• Advanced anomaly detection algorithm capable of discovering subtle, cross-channel abnormalities
• Iterative active learning approach for test plan optimization (Next Test Recommender)
• Unique visualization tool for optimization (Parallel Coordinates)
• Ability to reduce test steps by up to 70%
• Ability to find more than 90% of known issues in measurement data
• Ability to calibrate systems with 20x less driving hours and cut total test time by up to 20x
• Use of variational autoencoders for generative design and efficient geometry parametrization

Funding rounds we track

From the AI funding tracker — rounds as reported by the linked publications.