About
Analytics has a gap. We founded Xeplr to close it.
The Xeplr Pro Suite is one integrated analytical layer, from raw data to prediction. Xeplr OS is the foundation underneath it, published openly so smaller teams can build serious software too.
Why we founded Xeplr
Analytics has a gap, and it has been bleeding most companies. Data lives in one tool, the warehouse in another, reports in a third and predictions somewhere else. Every seam between them costs time, money and trust in the numbers — and no product on the market closes all of them.
Xeplr is an end-to-end analytical layer: model the data, aggregate it, report on it, predict from it and automate around it, in one integrated suite. It gives a company a strong analytical foundation, where millions of rows are no longer a problem.
Why the foundation is open
Smaller companies rarely get to build software the right way — there is never the time or the team to get authentication, jobs, access control and data handling right before the product itself. We want to change that, and help the open-source community build advanced software with ease.
Every data product is built on the same unglamorous pieces: authentication, a job runner, a schema primitive, a rules engine, a table, a chart. We wrote 22 of them, and there is no version of our business that is improved by keeping them private.
So they are MIT licensed and on npm. Use one, use all of them, or read them to decide whether you trust the applications built on top. What we sell is those applications — BI, the data warehouse, AutoML and Workflow — assembled, wired together and supported. Not the parts.
How it is put together
Each UI package splits model, controller and design, so presentation can be replaced without touching logic. Each service starts the same way and mounts into an Express app you already have. Nothing assumes it is the only thing running.
Where it is going
The data warehouse's aggregate-table engine is on DuckDB today and ClickHouse next — the same columnar, aggregate-state thinking either way, which makes the migration mostly mechanical. AutoML is the newest of the four, and the one where the shared model pays off most: features derived for a prediction go back into reports.
Founders
Closing the gap for large companies, and giving smaller ones the foundation to build on: that is the better world we set out to make.

Vikas Bhandari
Founder & CTO
Vikas sees problems differently. Where others see a product that is “good enough,” he sees what is missing — the friction users have learned to live with, the limitations nobody questions, and the possibilities that existing products overlook. He combines that vision with deep technical expertise to build solutions from the ground up. At Xeplr, Vikas is driven by one goal: to solve a problem that matters and build a product that is genuinely better, rather than simply another alternative.
LinkedIn ↗
Ankur Malhotra
Co-Founder & CEO
Ankur is a product builder and business thinker with experience at Paytm and Tokopedia. He has a rare ability to see what can make an idea matter, build it from the ground up, and turn it into something people actually want. Combining product instinct, market understanding and relentless execution, Ankur is driven by one thing — turning good ideas into products that create real value.
LinkedIn ↗Pradeep Paul
Co-Founder & COO
Pradeep Paul is a business strategist and operator with a strong instinct for identifying opportunities and turning them into real business outcomes. As COO, he brings commercial insight, strategic thinking, and execution discipline to the company — connecting market opportunities with the right people, processes, and decisions to drive sustainable growth.
See it against your own data
A demo runs on your schema, not a sample dataset. Thirty minutes, no deck.