Ashish Chatterjee.
Every bank, hospital and energy network is being asked to modernise decades of infrastructure and adopt artificial intelligence at the same time. The people who can actually do both are rarer than the headlines suggest.
There is a gap in the technology industry that very few people talk about honestly. On one side sit the announcements, the AI strategies, the transformation roadmaps presented at conferences. On the other sits the far less glamorous work of making any of it function inside organisations that cannot afford to fail. Banks that clear millions of payments a day. Energy networks that keep the gas flowing. Hospitals and health plans that process claims for millions of people.
The distance between those two worlds is where most transformation programmes go to die. Industry research has consistently put the failure rate of large digital transformation efforts at around seventy percent, and the reasons are rarely technical. Cloud platforms work. Microservices patterns are well documented. What breaks is the harder part, which is moving a complex organisation from one way of operating to another without disrupting the thing it exists to do.
A career built in that gap
Ashish Chatterjee has spent eighteen years working precisely in that space. His career spans banking, energy, healthcare and telecom across the United Kingdom, the United States and Europe, and the pattern across it is consistent. He is handed programmes where the tolerance for error is close to zero, and he delivers them.
Early in his career he led the consolidation of a global trading platform for an international investment bank, merging three separately maintained regional systems covering Europe, Asia Pacific and the Americas into a single codebase while migrating the architecture from a monolith to microservices. Unplanned downtime in a live trading environment can cost over a million dollars an hour per region. The programme completed without a single critical defect in production.
At Fiserv he directed enterprise platform delivery with zero major post release defects and introduced DevOps practices that cut operational effort by thirty percent. At John Deere Financial he ran an enterprise Agile Release Train that accelerated delivery cycles by forty percent and delivered the full committed business value within five months. At Wipro he spent over four years driving cloud and data transformation for the UK's gas transmission network, work that included modernising legacy analytics platforms and executing a large scale data migration during the separation of National Grid and National Gas.
More recently he delivered an Open Banking platform for a major UK bank, serving one of Britain's largest retail brands. The platform supports six to eight million registered users and handles three to five million transactions a day. It has run at 99.99 percent uptime since go live with zero defect releases, while meeting Open Banking UK standards, PSD2, FCA guidelines, GDPR and ISO 27001 at the same time.
Where the contribution shows
The clearest measure of contribution is what happens to an organisation after the programme ends. At TriZetto, Cognizant's healthcare digital engagement platform serving over two million consumers across more than twenty health plans in the United States, Chatterjee led an Agile and AI transformation. Delivery predictability improved, teams began surfacing problems early instead of escalating them late, and AI capabilities were introduced across member engagement, provider collaboration and claims processing. Following the changes, the platform recorded revenue growth of approximately 8.5 million dollars, around thirty percent over the period.
He is currently leading a large scale mortgage data migration for Nationwide Building Society as part of its integration with Virgin Money, one of the most significant consolidations in recent UK retail banking history.
What he sees coming
Ask him about artificial intelligence and the answer is noticeably less enthusiastic than the industry standard. Most financial and healthcare institutions, in his assessment, are considerably further from production ready AI than their public communications suggest. The data foundations are not in place. The governance frameworks covering model risk, explainability and accountability are still being written.
His view is that the work that matters over the next two years is not deployment at all. It is building the data infrastructure, the governance capability and the organisational discipline that make responsible deployment possible afterwards. He has applied that sequencing himself, auditing existing rule based systems and establishing data readiness before introducing models, and it is the approach he argues the sector needs to adopt more widely.
The broader vision he describes is one where technology leadership stops being purely an engineering function. As AI moves into regulated environments, the differentiating skill becomes the ability to hold together technical depth, regulatory literacy and organisational change at the same time. On his account, the institutions that will lead the next decade are not the ones deploying fastest. They are the ones doing the unglamorous groundwork now.
It is not a message designed to excite a conference hall. It is, on the evidence of the last eighteen years, the one more likely to be right.
