Company — ACE

Case Studies

01$4.5 Billion AUM PE Firm, Dubai

An operating system for the whole investment function

A Dubai-headquartered private equity firm with roughly $4.5B under management. Origination, diligence, portfolio and LP reporting each ran on their own tools, so the analysis a deal team does by hand was done again every time it was needed. We built a single connected surface across the firm’s investment activity: twelve modules running from pipeline CRM and deal scoring through due diligence, financial DD and a knowledge bank, into NDA, CIM and IC memo generation, and out to the portfolio dashboard and LP reporting. The analysis is generated from the deal record rather than assembled beside it — deal scores carry the fit breakdown behind them instead of a black-box number, DCF and LBO models run from WACC out to IRR and MOIC, and diligence checklists, risk registers and QoE reports come off the data room. Every generated artefact lands in an audit log: who, what, when, and from which inputs. It was delivered as a working, navigable product seeded with sixteen of the firm’s own holdings and twenty-one live GCC and India pipeline deals — each with KPIs, value drivers and an exit-readiness score, across healthcare, education, F&B, consumer, mobility, logistics and e-commerce — so the team evaluated it against deals they already knew rather than a demo dataset.

02$450M Buyout Fund

Rebuilding a fund’s institutional memory

A lower-middle-market buyout fund managing $450M across eleven portfolio companies in industrial and business services. The constraint was not strategy — it was knowledge fragmentation. Every deal was treated as an isolated episode, portfolio data ran 30 to 45 days behind, and exit preparation was rebuilt from scratch each time. We built six connected systems across the deal lifecycle, from deal intelligence and an investment-committee memory layer through to a portfolio performance data model and an exit narrative builder. At the seven-month mark of a fourteen-month engagement: IC preparation down 47%, the monthly reporting cycle compressed from eleven days to three, problem detection down from 45–60 days to 12–18, and buyer diligence questions at exit down from an average of 380 to 150.

03UK Family Office — Acquisitions & Private Credit

Taking the principal out of his own critical path

A single-family office running acquisitions, private credit and property, built around one principal who had exited his operating business. He was the constraint: sourcing, diligence and portfolio oversight all ran through him personally. We deployed nine systems against that — an off-market dealflow agent that emails targets inside his buy-box every morning; a bridging-finance engine that rescans millions of company filings weekly and returns roughly 5,000 qualified leads; a financial diligence agent now live on an active acquisition; and a company brain giving instant recall across every document, call and deal. Measured in July on live systems only, the engagement had returned 350 hours — a pace of around 890 hours a year, or £64,080 of analyst and advisory work at UK market rates.

Fulfillment Partner

We have partnered with ASZ Technologies who boast a team of 42 experienced developers and who have experience working with household names like Patek Philippe, Pfizer and NUS. ACE leverages the technical knowledge and domain expertise of ASZ to deploy enterprise grade services to our clientele.

ASZ has been building custom software since 2008, out of Bangalore, across application development, systems integration and IT consulting for multinational clients. That is a different discipline to the one most AI firms are staffed for — long-lived production systems, integration into estates that already exist, and delivery against fixed dates. It is also the discipline that separates a working demo from infrastructure a firm can run on.

The division of labour is deliberate. ACE holds the client relationship, the diagnosis, the architecture and the quality bar. ASZ supplies engineering capacity, drawn per build rather than carried as fixed headcount. Clients contract ACE and only ACE — the bench sits behind us, working to our method and under our review. It means a deployment can scale from a single agent to a full platform without us hiring against it, and without a client absorbing the risk of a firm learning to build as it goes.

Our Story

ACE started as a vision to bring clarity and excellence to the private capital industry surrounding the biggest technological innovation of our time — Artificial Intelligence. Back in 2022, when OpenAI launched ChatGPT, the world saw a technological leap in capability. This was the world’s first introduction to performance grade AI. Around this time our founder, Hamza, began experimenting with the new models extensively, and arrived at the conviction the firm was built on — soon businesses will be able to use this technology as a value creation lever.

That conviction met an obsession with private equity, its structure and its model. We found that financial engineering is not as reliable as it used to be and operational value creation was starting to emerge as a large factor in buy-side and sell-side decisions.

Once Anthropic came out with their models, the experiments moved into a real operating business — a family-owned bread manufacturing factory. We realised that there is immense value in building solutions internally with AI. These solutions kept getting better and cheaper with more usage by way of data collection and analysis. Around 2025 we began piloting that work with a few PE firms and investors, and eventually began onboarding clients.

While spending time deep in the real world, building with AI, we were struck by the sheer uncertainty and ambiguity around AI that plagues enterprises in today’s time. Excess token expenditure, broken agents, complex AI systems that don’t work — these were common problems in the lower-mid and mid-market that nobody was willing to solve.

This led to the birth of ACE. We’re brutally focused on Accelerating Company Excellence with AI.