AI consultancy.

An AI consultancy helps an organisation decide where AI (artificial intelligence) earns its place, then build and run it so it can be trusted. ddx covers each step: AI strategy and readiness, AI assistants grounded in your own data, and AI in how projects are delivered, proven on our own work first.

For leadership and product teams putting AI to work.

Request an AI Audit workshop

AI strategy

We define where AI earns its place in a product or an operation, what it must not do, and how its results will be measured.

Which model to use is usually the most discussed decision at the start, and the least consequential. What decides success is how the AI is grounded, governed and constrained. So the strategy starts with the job the AI is there to do, the information it can rely on, and the checks it must pass before anyone depends on it.

AI strategy sits in our Strategy practice, alongside the commercial model and the roadmap.

AI readiness

The AI you’ve approved isn’t always the AI you’re running.

Approved tools sit beside shadow AI: tools on no official list, fed with data nobody has mapped. In two weeks, the ddx AI Audit finds what AI your teams are running, shadow use included, where it leaves you exposed, and where it’s worth scaling.

It ends with a findings report, a list of opportunities scored by value and feasibility, and a six-month roadmap with owners. The roadmap is yours: run it in-house, or bring us in to help deliver it. To start, request a workshop.

Data foundations

An AI system can only use the data it can reach. In most organisations, that data sits in separate systems, in different formats, owned by different teams.

So before AI is built on it, the data needs foundations: shared data models, integration standards, secure interfaces between the systems that hold it, and clear rules on who owns it and who can see it.

AI assistants and implementation

For Torn, a browser game with more than 95,000 daily players and 20 years of lore, we built an AI assistant that answers inside the game’s own logic.

It runs on a leading large language model (LLM), built so the model can be swapped as better ones arrive. Retrieval-augmented generation (RAG) grounds each answer in verified game data: the assistant looks up the facts before it replies. A guardrail system sorts what each player is asking, blocks misuse and keeps answers in Torn’s voice, and more than 1,000 automated tests run before each deployment, so the model can be upgraded without a rebuild.

Retention rose 7% among players who used it. Read the Torn case study.

The same parts go into every AI assistant and large language model system we build: retrieval, guardrails and evaluation.

AI in delivery

AI changes how projects are run, as well as what they build.

ddx PMO reads the email, chat, work-tracker and document traffic a project already produces, and builds one live record of every risk, issue, decision and action, each linked back to the message it came from. Where the evidence isn’t in the text, the field stays blank, and anything uncertain waits for a person to approve it.

Health rolls up from project to programme to portfolio, and status reports assemble from the record, every line traceable. For programmes we run for clients, see PMO consultancy.

AI built into everything we do.

AI has been part of how ddx works since 2019, embedded in our everyday crafts. AI handles the volume, our team brings the judgement. Time saved goes back into the work: exploring more options, testing more assumptions and refining before we build.

More angles tested, more rounds of refinement, more rigour in the checks, on the same budget and the same clock.

  1. Our own first client

    ddx PMO, our AI-enabled PMO (project management office) platform, runs on ddx's own live engagements first, so by the time it reaches clients it's already proven on real work.

    ddx PMO
  2. A governed practice

    Behind it sits a governed practice, run to the PMI (Project Management Institute) Standard for AI in portfolio, programme and project management, with a documented playbook covering every team.

    ddx AI Audit
  3. The AI framework

    Every AI tool we use sits under one framework. A named person owns every output, and nothing AI produces reaches a client until they've approved it. Client data goes only into approved tools. Everyone at ddx is tested on the framework every year.

Questions.

What does an AI consultancy do?
It helps an organisation decide where AI is worth using, then build, govern and run it. ddx covers the strategy, an audit of the AI already in use, the systems themselves, and how projects are run.
How long does an AI readiness audit take?
Two weeks, scoped and time-boxed. It ends with a findings report, a list of opportunities scored by value and feasibility, and a six-month roadmap with owners.
Which AI model should we use?
It matters less than it seems. On Torn's assistant the model was built to be swapped as better ones arrived; what decided success was grounding, guardrails and testing.
Will you train AI on our data?
No. On an AI Audit we work under a non-disclosure agreement, train nothing on what we see, and delete or return your data when the engagement ends. On all our work, client data goes only into the AI tools our framework approves.

Who to talk to.

  1. Gareth Davies

    Gareth Davies

    Co-founder and Partner

    Leads technology at ddx: systems integration, AI and digital products, including One Bangkok’s smart city platform. Before ddx, led large-scale enterprise projects and technology companies. Former Chairman of the British Chamber of Commerce Thailand, and a director of the Thai Board of Trade from 2023 to 2025.

  2. Peter Sergeant

    Peter Sergeant

    Director of AI and Automation

    Pete leads ddx’s AI practice, taking LLM systems from prototype to production for clients: in-game NPCs, retrieval at the edge, autonomous negotiation agents. He has twenty-five years of hands-on engineering behind him, from the BBC’s early iPlayer team to CTO of Broadbean, and still writes code every day.

Start a conversation.

Whether you're starting something ambitious or restarting something that's lost momentum, we'd like to hear about it.

Get in touch