Who We Are

We build the science before we build the software.

Five Angstrom is an AI & Scientific Computing company — every product, model, and partnership on this site comes out of the same research group, working from first principles in physics, chemistry, and biology outward.

Most companies that call themselves "AI-powered" are software companies with a machine learning feature. Five Angstrom is the reverse: a research group that also happens to build software — one where the model, the simulation, or the algorithm is the product, and the interface around it is secondary.

Where we started

An academic research group, incorporated.

Five Angstrom was founded by researchers working in computational physics, computational biology, crystallography, and systems engineering — several of us out of the University of Madras. Before Five Angstrom existed as a company, this group was already publishing peer-reviewed research in conformational sampling and molecular modelling.

We're incubated at the MCC-MRF Innovation Park, Madras Christian College, Chennai — which is also where our research and engineering work happens day to day. The company exists to take that research further than a university lab typically can: into open-source tools, into hospitals, and into production systems that institutions actually run on.

What that means in practice

Why "AI & Scientific Computing," specifically.

We use that phrase deliberately, as two connected things rather than one buzzword. Scientific computing is the discipline: molecular dynamics, conformational sampling, force fields, statistical rigor — methods with decades of grounding in physics and chemistry. AI is where we've focused our own contribution: generative models that make those scientific methods faster and more capable, not a replacement for the rigor underneath them.

In practice, this shows up as work most "AI companies" don't do: publishing our methods in peer-reviewed journals, releasing tools as open source, and validating results against real experimental data — crystal structures, NMR ensembles, clinical imaging — rather than stopping at a demo.

How we work

Four things we hold ourselves to.

Publish the method

Our research is written up and submitted for peer review, not kept as an internal black box — our DNCS method is published in Physical Chemistry Chemical Physics.

Open source what we can

Tools built from that research, like DNCS 1.0, are released publicly under an open-source license for other researchers to use directly.

Validate against reality

Every method is checked against real crystal structures, NMR data, or clinical outcomes — not just internal benchmarks.

Apply it where it matters

Research becomes partnerships with real institutions — hospitals, dental colleges, and education platforms — not just papers.

Recognized as

Registered, and independently recognized.

Five Angstrom LLP is a DPIIT-registered startup, a corporate member of DSCI (the Data Security Council of India, a NASSCOM initiative), and registered under MSME — alongside our incubation at MCC-MRF Innovation Park.

Dr. Abraham Rebairo J.
Algorithm Design & Computational Physics
Dr. Roslin Elsa Varughese
Computational Science
Dr. Rishab Antosh
AI & Machine Learning
Mr. Mohandoss E
Database Management & Unix Engineering
Dr. Sam Paul D
Computational Biology

Bring us a hard scientific problem.

Whether it's a molecule to design, data to model, or infrastructure to secure — talk to the research group directly.

Get in touch