Skip to content
UCD Bots

UCD Bots

AI teammates for UCD teams.

An evidence-led, human-approved UCD pipeline — from research to a testable prototype — led by UCD Lead and run with specialist teammates on Grok Bot, who never log off.

Does not replace researchers or designers — the lead orchestrates; specialists draft; a person approves every gate.

Meet the AI teammates

  • UCD Lead

    Leads the pipeline across ten specialists — briefs by remit, synthesises stage packs, keeps evidence trails — always labelled draft, with a pause for you at every gate.

    View teammate

  • User Researcher

    Decision-linked questions, method, ethics and consent, interviews and usability tests, synthesis with insight strength.

    View teammate

  • UX Designer

    Information architecture, user flows, wireframes, and mid-fidelity structure. Progressive disclosure and one thing per page so complexity stays usable.

    View teammate

  • UX Planner

    Sequences discovery and design — RACI, research calendars, capacity risk, and a next step. Not a delivery manager.

    View teammate

  • Content Writer

    The words people use to finish a task — content models, headings, labels, microcopy, errors, help, and guidance.

    View teammate

  • Interaction Designer

    Flows, states, micro-interactions, transitions, feedback, and error recovery — inclusive by default.

    View teammate

  • Product Designer

    Research-lite through interaction and visual craft, with honest tradeoffs for business and engineering.

    View teammate

  • UI Designer

    Visual hierarchy on an agreed flow — proximity and type first, colour for meaning never alone.

    View teammate

  • Prototype Designer

    Chooses the right fidelity, then turns a UX planning handoff into prototype.md for Cursor, Codex, and POCs.

    View teammate

  • Business Analyst

    User stories, testable acceptance criteria, scope slices, dependencies, and residual risk — from the user and the outcome.

    View teammate

  • Accessibility Expert

    WCAG 2.2 AA criteria up front, in-flight reviews, and audits with named criteria, severity, and fixes.

    View teammate

From research to a testable prototype, with a person still in the loop.

What you can hold the work to

  • Human approval

    Work does not advance to the next stage until a person approves it.

  • WCAG 2.2 AA

    Accessibility is treated as a criterion in the method, not a later patch.

  • Evidence-led

    Claims stay tied to research, tests, and stated insight strength — not guesswork.

  • British English

    The teammates write in plain British English, matching the voice of the work.

  • Role boundaries

    Role boundaries follow NN/g · GDS · DDaT.

How the pipeline works

Typical path, not a gate order. The planner often starts early. The BA often runs alongside define. Accessibility is continuous. Prototype pairs with tests.

  1. Lead

    UCD Lead orchestrates discovery → live and routes craft to the specialists. You keep the approval; they keep the craft.

  2. Discover

    Decide what to learn, who is involved, and in what order. Planner and researcher often start here.

  3. Define

    Lock the words and the backlog around the user and the outcome. The BA often runs in parallel.

  4. Design

    Structure, behaviour, and visuals — with honest tradeoffs. Accessibility is in the method, not a later patch.

  5. Deliver and validate

    Prototype at the right fidelity and test inclusively. Accessibility can block “fix later.”

What Grok Bot is

Grok Bot is an always-on AI teammate platform from x.ai. Each teammate has its own compute, signs into a user's tools, and executes real work through the Grok Bot app.

This site is the public front door to a family of UCDOps teammates that run on that platform. It does not install software on this domain.