Managed neutronics and plasma simulation for labs and new nuclear suppliers
TL;DR Fluxhollow runs open-source nuclear and plasma simulation codes as a managed service, which means installed, validated and connected to compute, with an engineer on call. University groups and private firms entering nuclear supply chains pay a subscription plus project fees. It saves them months of set-up and gives them analyses they can defend.
Nuclear and fusion simulation in India sits mostly inside government institutions. The Institute for Plasma Research in Gandhinagar runs the ADITYA-U and SST-1 tokamaks and its own simulation work[1], and India supplies major parts of ITER, including the cryostat and in-vessel shielding[2]. Outside these centres, hardly anyone has the tools.
That gap is about to matter more. The SHANTI Act lets private companies manufacture nuclear equipment and take on some fuel fabrication[3]. A component maker bidding for small reactor work will need shielding, neutron transport and heat analyses. Today it has to hire scarce specialists, install open-source transport codes on its own machines[4] and validate them with no help at all.
Universities face the same set-up headache with less money.
What it does, in regulatory terms: engineering software and analysis support. It handles no nuclear material and needs no AERB licence. Check with DGFT whether any code or design data falls under export controls before serving foreign clients [Requires validation].
The seed proposed a collaboration platform for fusion research alone, and I've reframed it. India's fusion work sits in very few institutions, which is too small a base for a business. Widening to fission, shielding and plasma work brings in universities and the new private suppliers.
Fluxhollow offers a web workbench with open-source transport and multiphysics codes, ready-made nuclear data libraries and validated benchmark models. Users submit runs to cloud compute or to a national supercomputer allocation they already hold[5]. Each subscription includes hours with a staff engineer who reviews models and results. For firms, Fluxhollow also delivers fixed-scope analyses, such as a shielding study for a component, signed off by a senior reviewer.
University departments and research groups pay from project grants, as a yearly subscription per group. Private manufacturers pay per analysis project, with a subscription once they run their own models. I found no public price for managed nuclear simulation in India. Global cloud simulation platforms charge for compute and licences[6]. Fluxhollow's open-source stack removes licence fees, so the price is compute plus expertise.
The SHANTI Act took effect in December 2025, opening equipment manufacturing to private firms[3]. The Nuclear Energy Mission allocates ₹20,000 crore to small modular reactors and targets five running by 2033[7]. National supercomputing capacity has grown, with three PARAM Rudra machines dedicated in February 2025[5].
There is no public figure for nuclear simulation spending in India or for the number of private firms entering nuclear supply. The narrowest demand signal is policy-led - ₹20,000 crore for small reactors[7] will flow partly to component makers who need analyses. The adjacent figure is India's 9.1% share of ITER construction[2], which keeps a body of fusion engineering work in the country.
One US cloud-simulation platform sells with more than 1,350 preconfigured software packages to energy, aerospace and higher-education customers, but does not mention nuclear on its site[6]. The leading open-source code is free and community-built, with no paid support desk attached[4]. In India, the national supercomputing mission gives compute to academic groups[5] but not nuclear-specific set-up.
The gap is a ready, supported nuclear toolset.
| Difficulty | Moderate (37/100) |
| Starting capital | ₹40–80 L |
| Time to launch | 6–9 months |
The software is free. Making it reliable is not, because validating models against benchmarks, keeping data libraries current and handling hard support cases all need real nuclear engineers. Selling into universities is slow, and private demand depends on how fast reactor projects move.
No nuclear licence here. The SHANTI Act licenses plants, fuel work and radiation activities[7], and analysis software isn't among them, so keep it that way by handling no material.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't trust AI-generated results or reviews in place of a qualified engineer. A shielding study that goes to a client must be checked and signed by a senior reviewer.
Want to work out the AI layer for Fluxhollow? Talk to Yash →
I like this for a nuclear engineering PhD who has run these codes and wants to build a small, profitable firm. If you're expecting software-scale growth then it isn't for you. First prove that three groups or firms will pay for set-up and support within six months.
Plan for roughly ₹40–80 L, mostly for engineers who know the codes, cloud compute and building the workbench.
No nuclear licence, as long as you handle no material. Check SCOMET export controls with DGFT before you serve foreign clients.
Yes, universities and nuclear suppliers need this in many countries. Export controls on nuclear design data apply almost everywhere, so check yours first.
The free Fluxhollow validation kit has the line-by-line capital breakdown, the effort scoring, the export control and data checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for research groups and suppliers, and 5 assumptions to test. Download the kit →
Thinking about building Fluxhollow or something close to it? The make-or-break work is validated models that clients trust and the first paying supplier, and I'm happy to help you figure out how to land both. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.