A lot of what we build stays off the record — client NDAs, white-label deployments, systems embedded in products. This portfolio covers what we can share. If you want to understand the full scope of our work, the best conversation is a direct one.
AI Portfolio
Liquify AI Storefront - Shopify App Store
Our team has been building agentic commerce tools for Shopify merchants — AI-native apps designed to replace manual workflows and surface products the way customers actually think.
Beta
ShopAgent
AI-Powered Personal Shopping Assistant
ShopAgent lets users describe what they're looking for in plain language, then searches across Shopify merchants in real time to surface matching products with direct checkout links. Visual search, style onboarding, and a context engine that asks the right questions before it starts looking — built on Expo, Express/Node.js, OpenAI, and Shopify's Discovery API.
Expected release June 2026
AI Supplier Magic
Product Data Pipeline
Automates supplier data ingestion from end to end — ingesting CSVs, enriching product content with AI-generated descriptions and SEO copy, handling spelling convention conversion, and pushing clean data directly into Shopify. A real-time dashboard gives full visibility over every import. Currently in use across merchant stores; a SaaS release is in progress.
Production
Streamline AI
Embeddable Lead Intelligence
A rebrandable AI chat widget that replaces static lead capture forms with a structured conversation. Streamline gathers the context a business actually needs — qualifying leads, understanding requirements, and handling file and PDF intake — then automatically enriches each lead with company context and pushes the complete record directly to your CRM. Full-stack React and Express, configurable per deployment.
Production
Crazy Edge
Multi-Agent Ecommerce Suite — Sunset
A full suite of AI agents for ecommerce operators: Google Ads management, SEO and GEO/AEO optimisation, email and SMS, monthly deep audits, automated compliance monitoring, site testing, and brand clone detection. Seven products built and shipped under one platform. The market wasn't ready for the packaging — but the agent architecture, orchestration patterns, and what we learned about selling AI-as-a-service informed everything we've built since.
Sunset
AI Prediction Engine
Internal Research Project
A time-series prediction system built on AWS SageMaker, trained on historical stock price data and enriched with quarterly earnings reports. The pipeline ingests, aligns, and processes both data streams to generate short to medium term price direction signals. Built and used internally to explore what's achievable when you combine structured financial data with ML forecasting infrastructure — not productised, but a solid proof of concept for applied ML at scale.
Research
Content Marketing Pipeline
Internal Tooling
An internal system that automates the end-to-end content workflow for client marketing programmes. It handles keyword research, de-canonicalisation analysis, content gap planning, and generation — turning what's typically a fragmented, manual process into a repeatable pipeline. Used across client accounts to produce newsletters, blogs, and ongoing SEO content at scale.
Internal Tooling
Stack
Our solutions are built using leading technologies, including OpenAI, Claude, and Gemini. We are not affiliated with these companies; we simply leverage their tools to deliver the best results for our clients.
We are Cyber Essentials Certified, meeting UK government-backed standards for cyber security.