About · Adam Jasinski

About Adam London, UK

I’m Adam. An architect who builds, an engineer who thinks commercially, and an advisor who stays close to the work.

I help organisations connect the decisions made in the boardroom with the systems their teams can actually deliver. Enterprise AI is where those conversations come together.

A little about how I work

Three perspectives / One practice01—03

Different perspectives. Shared responsibility.

01 / My approach

Stay close
to the work.

The practical work changes the advice. The commercial context changes the architecture.

I build agent systems myself: writing Python, working with agent SDKs and MCP, and developing evaluation harnesses and tracing. It gives me a practical basis for setting technical direction with engineers and explaining difficult decisions to executives.

That perspective also comes from leading teams, shaping opportunities and owning delivery. A system has to work for the people operating it, the organisation investing in it and the teams responsible for its behaviour.

What I bring into the room

Make the decision
concrete.

Connect the ambition to investment choices, architecture constraints and a delivery path that people can act on.

Build the controls
into the system.

Make governance part of execution: observable behaviour, clear authority and evidence that supports the next decision.

Follow through
to the outcome.

Stay connected to engineering, adoption and value. Keep learning from what happens when the system meets the organisation.

02 / The journey

A career built
across boundaries.

From enterprise foundations to commercial leadership and agentic AI. Each chapter informs the next.

2012–2021

The foundations

Learning how enterprises work.

Consulting and architecture roles across HPE, financial services, government and cloud engineering. At Contino, I led an 18+ person team delivering a business-critical AWS platform.

2021–2025

Engineering meets commercial strategy

Taking responsibility for the bigger picture.

Technical direction, R&D leadership and proposition strategy. I was instrumental in a £20m acquisition, then worked across strategy at PwC and architecture and go-to-market at Merapar.

2025–now

The current chapter

Making agentic AI work at enterprise scale.

At Capgemini Invent, I work across agentic AI transformation, federated control planes and evaluation. Alongside that, I build the tools that let me explore these problems directly.

Explore the full career timeline

In progress

MSc Artificial Intelligence, University of Bath

Oct 2025present

Enterprise AI Architect, Capgemini Invent

Agentic AI architecture and transformation across financial services, public sector, telecoms and energy. Leading engineering teams and advising senior clients on platform strategy, governance and adoption.

Jan 2026present

Founder & Principal Contributor, AIMux

Open-source Agentic AI Service Mesh

Oct 2025May 2026

Board Advisor, A2A Net

Commercial and product positioning advisory for the full-stack A2A protocol platform.

Sep 2025Apr 2026

Board Advisor, Kora Intelligence

Product Strategy advisory for a marketing AI agent platform.

Oct 2023Sep 2025

Senior Architect & GTM Strategy Lead, Merapar

Service portfolio development, go-to-market strategy, technical oversight.

Apr 2023Dec 2025

Founder & Principal Advisor, Ashtons Capital Intelligence

AI-led due-diligence platform architecture, engineering and advisory.

Oct 2022Apr 2023

Strategist, PwC UK

Service portfolio development and integration, pre-sales and business value analysis.

Feb 2021Oct 2022

Technical Director & Digital Business Architect, Automation Logic

Led technical and value-proposition strategy through a £20m acquisition. Led a 25+ person R&D team, cutting customer time-to-value by 45% and reaching 74% profit margins.

20122021

Earlier consulting and architecture roles

Defence, financial services, and telecommunications.

2014

BSc Computer Science, Brunel University

A founder’s perspective

Through Ashtons Capital Intelligence, I architected and delivered an AI-led due-diligence platform. I also advised A2A Net and Kora Intelligence on positioning, commercial strategy and product validation during 2025–26.

Always learning

I’m studying for an MSc in Artificial Intelligence at the University of Bath. Earlier study includes Computer Science at Brunel and executive development at Yale and Oxford Saïd.

03 / On my desk

Questions I’m
building answers to.

Explore all work

CAESAR

How do we know an agent is doing the right thing?

Evaluation, trace analysis and safety gates. My MSc dissertation at the University of Bath.

Applied research

Metis

How do coding agents work well together?

The meta-harness that orchestrates AI coding agents - deterministic dispatch, scope as a contract, cross-vendor review, enforced by exit codes instead of hope.

In use by developers

AIMux