Services
AI consulting services for UK enterprises
Four core services cover moving one function from people-run to agent-run, governing the AI you already run, putting agents into production with a tested stop, and setting the rules for how your engineers build with AI. Each one, like the specialist services below, starts with a written scope.
Which service do you need?
Core services
Risk, security and compliance
AI is in production and you cannot yet show the evidence.
AI governance →Platform and engineering
You have an agent to put into a live system.
AI agent harnesses →CTOs and heads of engineering
Engineers already use AI to write code and you need common rules.
AI-enabled SDLC →COOs and function owners
One function still runs on people, and you want an agent to run defined work inside it, with a way back.
AI operations →
Specialist services
Energy and utility operations
You want to use AI in billing, metering or network operations in line with Ofgem's guidance.
Energy and utilities →Security leads
Agents can read untrusted content and act on your systems.
Agent security →Platform and finance
Model spend is growing and nobody can say what each use case costs.
LLM cost engineering →CTOs with legacy estates
A legacy system is costly to change, and you want to know whether agents can help migrate it safely.
Legacy modernisation →
| You | The problem | Service |
|---|---|---|
| Core services | ||
| Risk, security and compliance | AI is in production and you cannot yet show the evidence. | AI governance |
| Platform and engineering | You have an agent to put into a live system. | AI agent harnesses |
| CTOs and heads of engineering | Engineers already use AI to write code and you need common rules. | AI-enabled SDLC |
| COOs and function owners | One function still runs on people, and you want an agent to run defined work inside it, with a way back. | AI operations |
| Specialist services | ||
| Energy and utility operations | You want to use AI in billing, metering or network operations in line with Ofgem's guidance. | Energy and utilities |
| Security leads | Agents can read untrusted content and act on your systems. | Agent security |
| Platform and finance | Model spend is growing and nobody can say what each use case costs. | LLM cost engineering |
| CTOs with legacy estates | A legacy system is costly to change, and you want to know whether agents can help migrate it safely. | Legacy modernisation |
AI agents · EU AI Act
AI governance
Written controls for your AI agents and AI systems that your risk team can evidence.
- Agent identity, delegated authority
- Accountability, audit and security
- EU AI Act, UK regulatory guidance
- ISO/IEC 42001, NIST AI RMF
- Classification of each AI system
For risk, security and compliance leads governing AI agents and systems.
AI agents · Production
AI agent harnesses
Agents built into the systems and release process you run, with a tested way to stop them.
- Orchestration and tool permissions
- Evaluation, guardrails and logging
- Tested switch-off and rollback
- Automation behind existing interfaces
- Delivery via your release process
For platform and engineering leads putting agents into an existing estate.
Engineering · Quality gates
AI-enabled SDLC
How your engineering organisation builds software with AI, with a gate at every step.
- Policy for AI-assisted development
- Quality gates from spec to release
- Record of every AI-assisted change
- Rollout from pilot teams outward
- Works with any coding assistant
For CTOs and engineering leaders bringing AI into delivery across many teams.
One function · Controls first
AI operations
One business function, from the team that runs it to an agent that runs it inside written rules.
- Assessment of the function as it runs
- Decision rights, controls, rollback
- Pilot beside the current team
- Handover, or a managed run
- Fixed fees, then a capped outcome share
For leaders who want one function agent-run, with a way back.
Within AI governance
- EU AI Act
Map your AI to the Act, with evidence.
- ISO/IEC 42001
Gap assessment and readiness. We do not certify.
Specialist services
The same controls-first approach, applied to one sector or one problem.
Energy and utilities
AI for regulated energy and utility companies
Led by a former CTO of an energy-sector software company.
Agent security
AI agent security and red teaming
Testing what your agents can be made to do, and fixing it.
AI platform
AI platform and LLM cost engineering
Knowing what each use case costs, and whether a cheaper model would do.
Legacy modernisation
Legacy modernisation with AI agents
Capture what the system does first, then migrate in stages.
Related guideDoes the EU AI Act apply to UK companies?
How an engagement runs
Change to a critical system is slow, checked work: know the boundary, prove the new path, keep a way back.
Diagnose
We start from the system you run: its interfaces, data, release process and the cost of a wrong answer.
Design
We write the scope and the controls down first, including what is out of bounds and how it will be tested.
Embed
The work goes through your own release process, with your teams, and is checked against what was agreed.
Hand over
Your teams take ownership of what was built and agreed, and can run it without us.
Alex Vakhitov
Founder, Comonad · AI and software architect, London
Alex Vakhitov spent almost seven years as co-founder and CTO of Cloud KB, building cloud software for UK energy-market participants, in a regulated market. He founded Comonad in 2024.
Questions about Comonad's services
Which service should we start with?
The table above is a reasonable guide. Where more than one applies, start with the question your board or regulator is already asking: usually evidence for AI in production, sometimes the rules for engineers already using assistants. The introductory call is where we work this out with you.
Can the services be combined?
Yes. AI operations is often the frame: one function, controls first, with software delivery as the usual first function. Governance defines what a system must satisfy; the harness and the SDLC gates are where those controls are built and run. We scope each separately, so the advice and the build can each be reviewed on its own terms, and so you can stop after any one of them.
How does an engagement start?
Book an introductory call through the contact form, or email alex@comonad.co.uk, with a short description of the system or the engineering organisation. The first conversation is about scope. If we agree to work together, the first step is a written scope: what is in, what is out, and how the work will be checked.
Do you work with our legal advisers and auditors?
Yes. We work alongside your legal, risk and audit teams and prepare the records they review. We do not give legal advice: legal sign-off stays with your counsel. We do not issue ISO/IEC 42001 certification. Certification is carried out by certification bodies, and in the UK UKAS accredits them.
Are you vendor-neutral?
Yes. We use whatever models and tools your constraints allow, hosted or already in your estate, and we do not lock an engagement to one vendor. The harness, the evaluation and the controls are designed to survive a change of model.
Discuss an engagement
Tell us about the system and what you want AI to do in it. It can already be in production or about to ship.
Book an introductory call