From Contract to Quotation: Using AI on Your Live Data Sources
In our previous session, we showed how AI can turn a raw contract and a costing sheet into a draft quotation. Now we are going one step further: what does it take to make that tool work inside a real business, where data changes daily and every department has its own source of truth?
For our third Live AI Prototyping session, we are returning to the Quotation and Contract Intelligence problem, and rebuilding the prototype to integrate with dynamic, live workflows.
The first session showed how AI can ingest documents and generate a scope-aware quotation. This session shows what happens next. In most organisations, the data that feeds a quotation, pricing, labour rates, approved vendors, contract stipulations, does not sit in a static file. It lives in shared drives, updated by multiple teams, sometimes mid-deal. A prototype that cannot read that data in real time is a prototype that cannot be trusted.
We will architect the next version of this tool live in the room. You will watch us connect it to Google Drive and Google Sheets so it reads current data automatically, reflect updates made by other departments without manual re-upload, and generate quotations formatted to the client's own template, filed into the right folder in Drive.
We will also walk through deterministic AI methods, specifically graph-based, semantic, and hierarchical approaches, and explain when each one is the right call for keeping AI outputs accurate and auditable.
You will see us build a working prototype that:
Connects to live data: Reads directly from Google Drive and Google Sheets, picking up changes across departments in real time
Enforces logic: Automatically cross-references line items against current contract stipulations and approved pricing
Identifies variables: Flags out-of-scope items that require special approval, based on the latest data
Generates output: Produces a draft quotation formatted to your own template, saved into the correct folder in Drive
Agenda
You will receive the full agenda after registration. Light refreshments will be provided.
What we will show you
Dynamic workflow integration: How to connect an AI prototype to live data sources so it stays accurate as your business moves
Deterministic AI in practice: A practical look at graph-based, semantic, and hierarchical methods, and why they matter for commercial document generation
From prototype to something you can trust: The architecture decisions that separate a working demo from a tool your team will actually use
Who should attend?
This event is curated for business professionals, whether you are technical or not.
Curious business professionals: If you want to understand how AI tools connect to the systems your business already runs on, and what it takes to keep them accurate
Managers, directors and team leads: If you are thinking about how AI fits into your team's existing workflows, not as a standalone tool but as something that works with your data
Sales operations leaders: If you want quotation processes that reflect live pricing and contract terms, without someone manually checking every run
Bid managers and heads of tenders: If you are managing complex deals where the numbers shift and you need a system that keeps up
Solution architects and digital transformation leads: If you are scoping AI tools that need to integrate with existing data infrastructure and be trusted in production
About the Live AI Prototyping series
We are running this series to demystify what AI can and cannot do. We want to show you, as transparently as possible, what it looks like to build with AI, so you have a clearer picture of how to scope and start your own projects.
What is actually going to happen? We will build a live prototype to show you exactly how a raw business requirement becomes a working tool. We do not use black box platforms that hide the logic. We show you how to build a transparent, modular system that you own.
Interface with Next.js
Workflow manager with LangChain
Memory with ChromaDB
Live data with Google Drive and Google Sheets
AI model (Claude / OpenAI)
Limited seats to ensure quality interaction. Reserve your spot for a live engineering showcase.