contextric Book a call

Productivity by AI maturity

Productivity No AIAd hocTask-orientedagenticGoal-orientedagenticSoftwarefactories Most teams today

Engineering teams that get the most from AI.

For CTOs and engineering leaders. We look at how your teams build software and how they use AI to do it, show you where you stand, then help you move: with a clear plan, hands-on support, and the tools and agents built for you.

Book a free intro call See a recent review

An independent view from someone who has built it before.

Senior, hands-on experience of large programmes, from Critical National Infrastructure to FTSE 100 platforms, and of agentic engineering as it is practised today.

Engineering capability review

A clear, evidence-based picture of how your teams build and ship software today, compared with current practice, and a prioritised plan to improve it.

Delivery process
How work flows from idea to production: refinement, planning, build, test and release, and where it waits.
Architecture and code
Structure, quality, testing and technical debt, and whether they support where the business is going.
AI in engineering
The tools, models and agent workflows in use, what they cost, and what they actually deliver.
People and roles
How architects, engineers, analysts and testers work together, and the skills each role needs next.

How a review runs.

  1. Scope

    Agree the questions leadership wants answered, the teams in scope, and the access needed.

  2. Listen

    Interviews and working sessions across every role, from senior leaders to engineers, analysts and testers.

  3. Look at the work

    Repositories, pipelines, backlogs, AI tooling and spend: how work really flows, not how the process says it does.

  4. Compare

    Set against current industry practice, including what is working in agentic engineering right now.

  5. Report

    Findings and prioritised recommendations, presented to the leadership team. The report is yours to keep.

Legacy modernisation, delivered and assured.

Agentic AI can build almost anything, provided it knows what to build and has a way to check it got it right. Legacy modernisation is the rare case where both already exist: the system you're replacing is a working definition of right. We use it as an oracle, to write the specification, to build a safety net of tests, and to check every piece the agents rebuild.

Modernisation is rarely a straight copy, though. Most programmes also need to change how things work: new processes, retired features, cleaner data. So every behaviour is marked to keep, change or retire before it's rebuilt, and the safety net tests the changes as deliberately as everything else.

  1. Understand

    AI reads the code, data and documentation and maps what the system really does, including the rules nobody wrote down. The result is a specification people can check, with each behaviour marked to keep, change or retire.

  2. Safety net

    Before anything changes, we capture how the existing system behaves on real inputs and turn it into tests. Where the system and the specification disagree, people decide which is right.

  3. Agentic rebuild

    Agents rebuild one slice at a time from the specification, run the safety net and fix what fails, until the results match, or differ only where a change was intended. Then the slice goes live and that part of the old system retires.

Assuring it

Independent health checks for the programme's sponsor, one-off or at regular intervals.

  • Is the strategy right for this system: rewrite, replace in stages, or wrap and extend?
  • Is the plan realistic, and are the big risks owned: data migration, cutover, knowledge held by a few people?
  • Is AI being used safely on the code, and saving what it should?

Not just a report: we build it with you.

Recommendations only matter once they're delivered. We work alongside your teams to put them in place, and build what's needed.

  • Agent workflows for your engineers, with shared context, templates and guardrails
  • Migration and test tooling that does the heavy lifting on legacy code
  • Internal assistants that know your codebase, standards and architecture
  • AI agents that run business processes, from credit control to tenders
Talk to us about a build
Client
Global systems integrator
Sector
Public sector, legacy modernisation
Scale
6 projects, around 150 people
Role
Independent reviewer, reporting to the leadership team
Duration
Six weeks

Agentic engineering review for a global systems integrator

The client was modernising legacy systems for several public sector customers, and building its own AI engineering capability alongside. Each project had adopted AI tools in its own way, costs were rising, and leadership wanted to know what good looked like and how far they were from it.

What we did

We worked with people in every role, from architects to engineers, business analysts and testers, sitting in on ceremonies and following real pieces of work from requirement to release.

We looked at how each team used AI: which tools and models, what context they gave them, and what came back. Then we compared it with current practice in agentic engineering.

What we found

  • The most capable, most expensive models were used for every task, including simple ones
  • Agents were given whole repositories as context, so most tokens went on code that didn't matter
  • Each project had built its own prompts and workflows, with no way to share what worked
  • Analysts and testers were largely left out, so gains in coding were lost further down the line
  • Nobody was measuring what AI delivered, only what it cost
lower token spend, pilot teams
40%
more delivered per sprint, pilot teams
25%
projects reviewed
6

Outcome

  • A report for the leadership team with prioritised recommendations: models matched to tasks, scoped context, a shared playbook, analysts and testers in the workflow, and simple measures of value
  • Pilot teams cut token spend significantly while delivering more
  • The recommendations became the basis of the organisation's AI engineering standards

Talk to us about your engineering teams.

A 30-minute call to understand what you need answered. If a review or a build would help, we'll follow up with a short proposal: scope, approach, timing and a fixed price.

Choose a time

Or email hello@contextric.com