Skip to content

We build AI for work that has to be checked.

SupaMakers is a founder-led AI development company in London. We start with your own files, tools and awkward cases, build the smallest system that helps, and link every result to its source. You work directly with the two people making the product and technical decisions.

Who you work with

Shilpa Pawar

Founder

Founder and primary contact for new SupaMakers work, from the first workflow conversation through project scoping and delivery.

What we build

AI systems for the document work professional teams already do: reading and reviewing documents, pulling out the details that matter, drafting a first pass, and routing it to the person who decides. Each one is built around a single workflow and tested on your own material.

AutoBOQ is one example: it turns construction drawings into a bill of quantities an estimator can check. The same pattern applies to contracts, tenders, claims and compliance evidence, wherever a result has to be accurate and easy to review.

How we work

Start with the evidence inside the workflow.

A useful engagement begins with the actual material: drawings, contracts, tenders, claims, reports, spreadsheets, or system records. We map who owns the work, what a correct output looks like, where exceptions appear, and which decisions must remain with a person.

That evidence determines whether the next step should be a strategy assessment, a narrow prototype, a production agent, or better data infrastructure. It also gives the project a testable definition of success before more software is built.

What you can inspect

Our own products are the evidence.

Law AajKal, CasePulse, AutoBOQ and ContractScanner each show part of the same practice: document pipelines that keep their sources, AI output a professional can review, search across large archives, and clear limits on what the software decides.

Each build note states the evidence and the constraints. We do not add anonymous testimonials, client logos, outcome statistics or credentials that cannot be verified.

Read the build notes

AutoBOQ

Prototype
AutoBOQ turning a set of house drawings into a structured bill of quantities, grouped by section with units, quantities and drawing references

Drawing PDFs in, a bill of quantities an estimator can review out.

Limited-access prototype; codes on request from the tool page.

CasePulse

Live
CasePulse overview: judgments indexed, citation connections, judgments by court and the most cited cases

UK case law, searchable, with the citations between judgments linked.

56,000+ judgments and 411,000+ citation links: corpus counts from The National Archives' Find Case Law.

ContractScanner

Prototype
ContractScanner: a free, private AI contract review tool powered by an in-browser model

Contract triage that runs in the browser; the text never leaves the device.

The model runs locally via WebGPU after you choose to load it. No contract text is posted to a SupaMakers server.

Publishing and corrections

SupaMakers publishes technical guidance and courses alongside commercial services. Sources, authorship, AI assistance, regulated-topic limitations, updates, and corrections are governed by a public editorial policy. If a claim is wrong or a source has changed, send us the passage and a better source.

Read the editorial policy

Start with one workflow.

Tell us which piece of work you’d like AI to take on. We’ll reply with what a system could do, what it should leave to a person, and what a first version would take to build.