AI and Machine Learning
Retrieval, vector search and LLM features built to hold up under real traffic, not just in a demo.
Product engineering · Ahmedabad, India
Lotus Tech Labs is a small senior team building web and mobile products - and the AI features inside them - for companies in the US, UK, EU and Australia. No juniors hidden behind an account manager.
The work we lead with is AI and LLM product engineering - usually a feature that demos well and then falls over on real traffic. Web and mobile are how that feature reaches anyone, so we build those too, with the same people.
Retrieval, vector search and LLM features built to hold up under real traffic, not just in a demo.
Websites and web applications on React, Next.js and NestJS, built to stay fast as they grow.
Cross-platform and native applications for Android and iOS, built for performance and reach.
We are a small engineering team in Ahmedabad, building web and mobile products since 2024. Every project is worked on by senior engineers - nothing is handed to a trainee, and there is no account manager between you and the people writing the code. If we are not the right fit for what you are building, we will say so early.
More about how we workThe questions that come up on a first call, answered before you have to ask. If yours is not here, ask it on the call.
All four of us. We take one project at a time, so there is no pool your work gets assigned out of and nobody in between you and the build. The person who scopes your project is the person who writes it.
A discovery call first, where we ask questions until we understand the problem rather than taking notes to relay. Then a written PRD that you approve before any code exists. Then agile iterations where you see working software as it lands, and a UAT phase at the end where you sign off against what the PRD said.
At least one call a week, and a shared Slack or Google Chat channel for everything between calls. You are talking to engineers in both, so a question about a trade-off gets an answer instead of a status update.
TypeScript on Node: NestJS on the server, React and Next.js in the browser. Flutter and React Native for mobile. Python and its ecosystem where the work is machine learning or retrieval. We choose from that set rather than learning something new on your budget.
Yes. Some work is a defined scope with a finish line and some is an ongoing build, and we are set up for both. What we insist on either way is the written PRD at the start, because an open-ended timeline and an unclear scope together is how projects go wrong.
We are in Ahmedabad, India, working with founders and product teams in the US, UK, Europe and Australia. The weekly call is scheduled in your working hours rather than ours.
Staff augmentation: we take the problem, not a seat on your board. WordPress and PHP. And we do not run more than one project at a time, so if we are booked we will tell you rather than stretch.
It depends on the scope, and we would rather tell you honestly on a call than publish a number that is wrong for your project. The discovery call and the PRD that comes out of it are what turn it into a real figure, and the PRD is yours to keep either way.
There is more on how we work and what we have shipped, or go straight to booking a discovery call.
A couple of paragraphs is enough to start. You will get a reply from an engineer, and if we are not the right fit we will say so early.