Self-Service AI Go beyond the chatbot
Build your own AI agents to automate the financial, operational and reporting work you do today - grounded in your data, your processes and your ways of working. No code, and no waiting on Data/IT teams or needing to hire into these roles.
The shared context that lets your people and your agents speak the same language: how your business actually works, the rules you follow, how you define your numbers. Captured once in a format both humans and agents can read - so your agents give consistent answers instead of guessing.
One common home for your data - brought in from your ERPs, CRMs and spreadsheets into a platform like Microsoft Fabric. The same brain knows which data each process needs and how to shape it, so your agents always work from trusted, standardised data.
The result? AI agents you can trust - ones that respect your security rules, work only from data you've approved, and know enough about your business to give consistent, reliable answers instead of making things up.
Build your first custom agent skill - directly from Excel!
Excel to Agents
Why 95% of AI Projects Fail*
and why Your company won't be a part of this statistic
* Source: MIT report: 95% of generative AI pilots at companies are failing (Sheryl Estrada - Senior Writer and author of CFO Daily)
Framework
A five-step, practitioner’s guide – written for the Institute of Chartered Accountants in the UK (ICAEW) – to automating finance processes with AI agents instead of rigid macros. Covering the data foundation agents need, how to design an orchestrated team of sub-agents, and how to build, test and gain trust in the solution. All using Excel, Power BI, Fabric and Copilot Studio for a deterministic grounding of data and logic.
If any of this sounds familiar, keep reading.
"We bought Fabric. We bought Copilot. The board still asks what we’re doing with AI."
"Every team is building their own GPTs and agents. Nobody knows where the context lives."
"The vendor agents don’t understand our chart of accounts. The consultants want a twelve-month contract – with no guarantee we’ll be able to use AI the way we want anyway."
Self-Service AI - the architecture, in one diagram
Self-Service BI promised to free business teams from IT - and never quite delivered, because the tools were simply too hard to learn on top of a day job. Self-Service AI is the second attempt, this time with AI as your partner in the build. It is not a vendor product or a consulting engagement; it is a framework and architecture you stand up on the Microsoft stack you already own. Five context layers feed a governed agent builder your teams use to assemble agents for their own processes - drawing on the same context, the same data, and the same governance every time.
Microsoft Fabric is the data layer. Copilot Studio is the agent builder. The five-layer context foundation is what makes everything in between actually work for your business.
A Chartered Accountant and Microsoft MVP (technology influencer/speaker with direct access to product groups at Microsoft) with an Executive MBA (Hons) and a first class degree from the London School of Economics.
Currently leads Data & AI strategic projects at Avanade (Accenture-Microsoft JV), focusing on Power BI and Fabric.
With 20+ years experience across Deloitte, KPMG, HSBC, Barclays and Accenture, Rishi combines deep financial expertise with cutting-edge data and AI capabilities to drive business transformation.
Built and taught by someone in the room every week.
Not theory from the sidelines - recognised expertise, applied in the field.
Microsoft MVP
Chartered Accountant
Executive MBA
FabCon Europe
SQLBits
Accountex/FD Show
In their words
What people take away.
"The event was wonderfully presented and segmented. The material was up to date and very relevant to my role."
Senior Finance Professional
LDI EVENT ATTENDEE (STRATEGIC INTELLIGENCE FORUM) · Q1 2026
I really enjoyed learning about AI agents – how I can explore and start to implement them in my workplace.
Senior Controller, Group FP&A
LDI EVENT ATTENDEE + POWER BI FOR FINANCE COURSE · 2026
Excellent networking. Great to see the level of engagement and the kind of people you were able to attract.
C-Suite Executive
LDI EVENT ATTENDEE (STRATEGIC INTELLIGENCE FORUM) · Q1 2026
Five layers. One context. Every agent.
Microsoft Fabric OneLake
Your governed business data. Tables, schemas, semantic models. One source of truth.
Validated metric definitions
Net revenue, working day, period close — the way your business defines them. Agreed once, reused everywhere.
MCP-served connections
Schemas and permissions for your Enterprise applications (such as SAP, NetSuite, D365 and HubSpot). Agents reach source systems through governed tools.
Reusable + custom skills
Utility skills (Excel, Power BI, Fabric) and custom business-process skills. Each one references the layers above.
The control layer
Permissions, agent policies and corporate controls — turning "anyone can build an agent" into "anyone can build a governed agent".
The single next step
Become a member. It is free, and it is the start of everything.
What you get
The outcomes our members are building toward.
1
Processes that run themselves - and prove it.
Automated month-end and variance workflows generate board-ready Power BI reports, formatted to your standard and delivered straight from Teams. No manual formatting, no copy-paste.
2
Answers, not dashboards.
A plain-English Q&A layer over your Fabric data – ask “what’s driving the March gap?” and get a trusted answer instantly. One agreed definition of every number, consistent everywhere.
3
A credible AI story for the board.
Specific, anchored in real processes, built on real data and real insights. Not another pilot deck or set of vendor promises.
4
A capability - and a team - your business owns.
Your own team of agents – business analyst, Power BI developer, data engineer, tester – building on a governed foundation you control. Not rented from a vendor, not trapped in a consultant’s laptop, and growing with every process you automate.
From zero to governed AI
How it works
1
Start here
Get free membership
One form, under 60 seconds. Instant access to the recording and the interactive Excel-skills tutorial – with e-learning courses and your data & AI readiness roadmap coming soon.
2
Know where you are
Take the AI Readiness Roadmap (coming soon)
A short diagnostic across all five context layers. See exactly where your foundation is strong and where the gaps are.
3
Learn and build
Follow your learning path
Courses, community calls and member resources – structured around your role, whether you’re leading, analysing or building with AI.
4
Ship something real
Build at the quarterly workshop
A hands-on build day. You leave with a working governed agent on your own data – not a prototype, not a demo.
Questions
Things people ask before they join
Yes – core LDI membership is free, with no time limit and no hidden upsell to unlock the essentials. I’d simply rather have practitioners in the room who genuinely want to build than chase sign-ups.
Yes. There’s a dedicated path for non-technical leaders – CFOs and senior commercial roles – who need to understand and commission Self-Service AI without writing a line of code, alongside deeper paths for finance analysts (starting from Excel and Power BI) and data practitioners. All three begin from first principles, so you join at your level.
This is the heart of why Self-Service AI is built on the Microsoft ecosystem. Your data never leaves your own environment: it stays in your organisation’s Microsoft tenant and Fabric capacity, in the region you choose – with multi-geo support, so different datasets can be held in different regions to meet local requirements.
Access is governed end to end through Microsoft Entra (identity, conditional access and permissions), so an agent only ever sees what the person using it is allowed to see. You’re not shipping your finance data off to a third-party AI provider – the intelligence comes to your data, inside your tenant, under the security and compliance controls you already run.
Microsoft Fabric is needed to run the LDI agent builder itself, though for workshops and demos I can set you up on a temporary account in my tenant. Each course also states exactly which tools and licences it needs, so you always know before you start.
It depends how much context you’ve built into your agent builder before you start. The quarterly build workshop is designed to get you to a working example in a single day – something your team can actually use, not a slide deck about what you might build.
Consultants build things for you; LDI teaches your team to build for themselves. The quarterly workshop is structured so your team owns every part of the agent they build – the data connections, the logic, the governance rules. The goal is a capability your business controls, not a dependency.
