Skip to main content

One team.
We build the app
and the automation.

You need software people can use, and the AI that makes it run. We ship both, and we run five products of our own.

Built and running:notetube.aikraftr.studiogetpullsy.comgymfast.appsigmacpm.com

products we build and run ourselves
5
products we build and run ourselves
systems shipped, all on the work page
9
systems shipped, all on the work page
years of founder experience in AI and backends
8+
years of founder experience in AI and backends
minimum project, MVPs in 3 to 5 weeks
$1,500
minimum project, MVPs in 3 to 5 weeks

Selected work

We build products, then we run them.

Running our own software in production is how we learned what breaks. You get that, not a first attempt.

  1. 01Our product

    NoteTube

    A lecture becomes a study system: chat, flashcards, quizzes, all grounded in the source.

    An AI learning platform that turns PDFs, YouTube lectures, recordings, and notes into a structured notebook with grounded chat, spaced-repetition flashcards, and graded quizzes.

    Web · iOS · Android · Chrome extension · MCP server

  2. 02Our product

    Kraftr

    One flat-lay photo becomes catalog-ready 4K on-model imagery.

    An AI catalog studio for fashion and jewelry brands. Upload one flat-lay or hanger shot per SKU and get on-model imagery at up to 4K, in ten aspect ratios, sized for Amazon, Myntra, Flipkart, and Shopify.

    Web · iOS · Android

  3. 03Case study

    Acasa Sentinel

    Give it a number plate. It returns the vehicle's route across a CCTV network.

    A vendor-neutral video analytics system built for the Gujarat Police Hackathon Innovation Challenge 2026. It reads number plates, re-identifies vehicles by appearance across cameras, reconstructs routes, and raises watchlist alerts for an operator to confirm.

    Operator console (web) · API

All work and case studies

Our specialty

AI that answers from your documents, and shows its source.

Retrieval-augmented generation is most of the AI work worth doing in a company. It is also where most demos fall apart. Here is what we do differently.

Chunk by structure
Sections, tables, and clauses, not every 500 characters.
Search two ways
Keywords and vectors together, then a reranker picks the best few.
Cite or decline
Every answer points at its passage. No passage, no answer.
Measure it
An evaluation set from your real questions, run on every change.
retrieval traceillustrative
Query

What is our notice period to terminate the vendor contract?

Retrieve

hybrid search · 40 candidates → rerank → 3 · access-filtered

  1. [1]

    MSA_2024.pdfp.14 · §11.20.91

    Either party may terminate this Agreement for convenience on sixty (60) days' written notice.

  2. [2]

    Vendor_Policy.docx§4.30.84

    Any notice period shorter than thirty (30) days requires written approval from the CFO.

  3. [3]

    Amendment_03.pdfp.20.79

    Section 11.2 is amended to replace “sixty (60) days” with “ninety (90) days”.

Answer

Ninety days. The master agreement sets sixty days[1], but Amendment 3 extends it to ninety[3]. Ending it on shorter notice needs CFO approval[2].

Who this is for

Operators who need it built, not workshopped.

You run a mid-size company. Work still lives in inboxes, spreadsheets, and tools that do not talk to each other. You do not have an ML team. You want a product people can use, and the automation that takes the busywork out of it.

About the team
The Acasa Labs team
A small team in Surat. One group, both jobs.

How we work

Build, then keep building.

  1. 01

    Sit with the operation

    We watch how the work actually happens. Inboxes, tools, handoffs. No workshop theater.

  2. 02

    Ship the product

    A working web or mobile app. Something your people can open on Monday, not a prototype graveyard.

  3. 03

    Wire the automation

    Agents and workflows on the product and the systems you already have.

  4. 04

    Stay on it

    We do not vanish after launch. The operation changes. The software should too.

Pritesh Kanani

Who you will work with

Pritesh Kanani, founder

Pritesh Kanani is the founder of Acasa Labs and a full-stack and AI engineer with more than eight years of experience building AI-powered applications and data-intensive systems. He specialises in LLM integration, retrieval-augmented generation, and Python backends, and takes products from zero to production while balancing performance, accuracy, and cost. He has worked in India and Canada, and is based in Surat.

Questions

What people ask first.

More questions
What does Acasa Labs do?

We are a software studio in Surat, India. One team builds your website, web app, or iOS and Android app, builds the AI and automation inside it, and then scales it after launch: speed, reliability, cost, and search. Our specialty is RAG, AI that answers from your own documents and cites the source.

How much does a project cost?

A website or catalogue site is $1,500 to $3,000. A mobile app MVP for iOS and Android is $2,000 to $5,000, and a web app or SaaS MVP is $4,000 to $8,000. An AI pilot is $5,000 to $10,000. A dedicated developer is $1,500 to $2,000 a month, and maintenance is $100 to $300 a month. The minimum project is $1,500.

How long does it take?

A website usually takes 1 to 2 weeks, and an MVP web or mobile app 3 to 5 weeks. Larger products are planned in stages, each with a demo and a finish line, and you see progress every week.

Who owns the code?

You do, from the first day. Repositories, cloud accounts, app store accounts, and credentials are in your organization, and we document how to run everything.

We are not in India. How does working together go?

We set our working day around yours, with at least 4 hours of overlap and a daily call for US clients in their morning. In Gujarat and Mumbai we meet in person, and we speak English, Gujarati, and Hindi.

Can we see something you have built?

Yes. NoteTube, Kraftr, Pullsy, and GymFast are live products of ours you can open today, and SigmaCPM is in early access. The work page shows each one, plus client work and case studies, with the stack and the engineering behind it.