AI, ML & Data Science

Capability

Autonomous assistants that plan, call tools and complete multi-step tasks across your systems, with human oversight where it matters.

Anthropic's Claude models, accessed through the Claude API, for assistants, document analysis, coding help and agent workflows that need careful reasoning.

A lightweight open-source embedding database that is quick to set up, ideal for prototypes and smaller RAG applications.

Capability

Image and video analysis such as object detection, OCR, quality inspection and face or scene recognition for products and operations.

Capability

Features that create text, images, code or summaries with large language and diffusion models, integrated safely into your products and workflows.

Google's Gemini models, available through the Gemini API and Vertex AI, for multimodal text, image, audio and video understanding.

An ecosystem of open models, datasets and libraries such as Transformers for fine-tuning and serving language, vision and speech models.

A fast language for scientific and numerical computing, suited to simulations, optimisation and data-heavy research workloads.

A high-level deep learning API that makes building and training neural networks faster and clearer, running on TensorFlow and other backends.

A framework for building applications on large language models, with building blocks for chains, agents, tool use and retrieval.

Meta's family of open-weight large language models, suited to self-hosted or fine-tuned deployments where data must stay in your environment.

A data framework that connects large language models to your documents and databases for accurate search, question answering and summarisation.

A numerical computing environment used in engineering and science for signal processing, simulation, control systems and algorithm prototyping.

Open-weight and hosted language models from Mistral AI, a strong option for efficient deployments where cost and control matter.

Capability

Packaging, deploying, monitoring and retraining machine learning models in production with versioning, pipelines and reliable infrastructure.

Capability

Text understanding for classification, sentiment analysis, entity extraction, translation and search, from traditional pipelines to modern transformer models.

The core Python library for fast numerical computing on arrays and matrices, underpinning scientific, analytics and machine learning code.

A tool for running open-source language models locally or on your own servers, useful for private prototypes and offline AI features.

OpenAI's hosted models, including the GPT family, for text generation, embeddings, vision and speech features inside your software.

A Python library for cleaning, transforming and analysing tabular data, the foundation of most data pipelines and analytics work.

A PostgreSQL extension that stores and searches vector embeddings, letting you add semantic search and RAG without a separate database.

A fully managed vector database for similarity search over embeddings, suited to production AI search and retrieval at scale.

Capability

Forecasting and scoring models for demand, churn, pricing, risk and maintenance, built on your historical data and explained in business terms.

A readable general-purpose language that fits backend services, automation, data processing and most machine learning work.

An open-source deep learning framework for building, training and deploying custom neural networks, from research prototypes to production models.

An open-source vector database with rich filtering, available self-hosted or as a managed cloud for semantic search and recommendations.

A language built for statistics and data analysis, a good fit for research-grade modelling, reporting and visualisation.

Capability

Personalised product, content and offer recommendations built from user behaviour and catalogue data to increase engagement and sales.

Capability

Grounding language models in your own documents and data so answers are accurate, current and traceable to their sources.

The standard Python library for classical machine learning, ideal for classification, regression, clustering and forecasting on structured business data.

Google's open-source machine learning framework, suited to production-scale training and deployment across servers, browsers and mobile devices.

A high-throughput inference server for open-source language models, used when self-hosted AI needs to serve many users efficiently.

An open-source vector database with built-in vectorisation modules and hybrid search, suited to knowledge bases and AI-powered search.