From AI Pilots to Operating Reality

Client:
Xceptor

Industry:
Financial Technology / Data Automation

Client presenter: Michael Kinloch, SVP Engineering, Xceptor 
Forte Group lead: Alex Lukashevich, Chief AI Officer

Presented publicly at: CTO Craft Conference, 10 March 2026

Watch the full on-stage recording

In Brief

Most organisations talk about AI in software delivery. Our client Xceptor is doing it across every phase of their product lifecycle, in production, with numbers to back it up. Working with Forte Group, they moved from ad-hoc AI tool adoption to a fully embedded, three-stage AI delivery model in under twelve months. This is not a pilot. It is not a proof of concept. It is how Xceptor builds software today.

The Challenge

Xceptor empowers business users within financial institutions to build automated processes that deliver trusted data. Their platform connects upstream and downstream data sources, extracts files of any format or structure, automates manual processing journeys across the full trade lifecycle, and provides a live web interface for business users to manage exceptions and operate what they’ve built. Purpose-built products span reconciliations, post-trade operations, tax processing, and client lifecycle management.

As a high-growth, PE-backed FinTech, Xceptor faced the challenge shared by most technology organisations in 2025: they had invested in AI tools and seen individuals move faster, but delivery velocity at the team level remained unchanged. The queue was the same. Releases were the same. The 5× and 10× productivity gains being promised by the market were not materialising.

The root cause was a "tool-first" approach, implementing tools into systems not yet prepared for them. To unlock true gains, Xceptor needed to understand their processes first and then apply AI to workflows ready to receive them.

‍

The Approach

A Deliberate Three-Stage Journey

Think about it like any other problem: break it down, start small, look for where you can have a lot of value with not a lot of effort. Forte and Xceptor executed a sequenced, measurable journey across three distinct stages to move from ad-hoc adoption to a fully embedded AI delivery model in under twelve months.

How Roles Transform

From Doing to Directing

The AI-native PDLC changes what people spend their time on. Roles do not disappear, they elevate. These roles are starting to combine over time. We’re going to see fewer roles but more specialist skills. Our teams are going to shrink, but we’re going to get more done with less.

“We were trying to get AI to build a process on our platform. When we gave it too much context, it did a terrible job. When we gave it just the bare minimum, a codebase link, the documentation, one good working example — within a day it was building an end-to-end process like our business user would.”

— Michael Kinloch, SVP Engineering, at Xceptor

‍

The Vision

Stages one and two are complete. The third stage, the fully AI-native PDLC , is being proven in production now. The goal: a full connector built from requirement to production-ready in under one day.

Where previously a connector might have taken multiple weeks to build, Xceptor is midway through a six-week MVP that targets sub-day delivery. The proof is a full agent ecosystem: a single orchestrator agent receives a feature request and coordinates parallel specialist sub-agents across product ownership, development, QA, and DevOps simultaneously. Every output passes through a human review gate before promotion.

‍

“We’ve put together a vision that we’ve been working through and building with Forte Group around our AI-native product development lifecycle. It’s based around a central agent ecosystem — how we get agents driving the product development lifecycle with humans in the loop for governance.”

— Michael Kinloch, SVP Engineering at Xceptor

Key Learnings & Evolution

The shift to an AI-native PDLC has elevated roles from "doing" to "directing."

• Product Owners now steer agent-built requirements rather than writing them from scratch.

• Developers focus on architecture and reviewing PRs drafted by AI.

• QA and Ops approve AI-generated suites and insights, focusing on high-level risk and strategy.

‍

Results

AI is now embedded and delivering measurable production results across all five phases of Xceptor’s PDLC.

Design: 83% Faster Prototyping

The UX team now uses AI to generate all design prototypes, allowing for more variations and better solution validation before engineering begins. This has returned 32 days per designer, per year.

Requirements: 3x Faster Story Creation

By using AI to generate first-draft requirements from meeting transcripts, story creation time dropped from over three hours to under one hour. Additionally, the rework rate was slashed from 30% to below 10%.

Build: 85% Engineer Adoption

Through a champions program and role-specific playbooks, Xceptor achieved widespread practitioner-led adoption. Engineers generated 18,064 AI events in just six months.

Test: 75% Less Scripting Time

By switching to Claude Opus and the AI-native Playwright MCP framework, Xceptor reduced test scripting time by 75%. Now, one QA engineer produces the equivalent output of four.

Production: Zero Missed P1 Incidents

Forte built an autonomous AI agent that scans logs across 170+ customer instances. It triages exceptions and proposes fixes, reducing triage time from days to hours while ensuring zero missed P1 incidents.