---
title: Riemann Computing Launches LonScript, a Scientific Computing Language It Says Runs 1,434 Times Faster Than Python
description: Riemann Computing unveils LonScript, a scientific computing language it says runs 1,434x faster than Python, as the Troy, Michigan firm raises on Wefunder.
author: Darie Nani (Editor-in-Chief)
updated: 2026-08-12T08:16:59.718Z
canonical: https://www.sovereignmagazine.com/article/riemann-computing-lonscript-1434x-python
image: https://cdn.nanimediahouse.com/pexels-vivid-blurred-close-up-of-colorful-code-on-a-screen-represen-2004161.jpg
categories: Science &amp; Tech, Artificial Intelligence
content_type: Spotlight
region: Michigan
publication: Sovereign Magazine
about:
  - type: Organization
    name: Riemann Computing
    description: Riemann Computing Inc. is a Troy, Michigan technology company founded by Andrew Kamal in 2018, originally as Stark Drones Corporation. It develops high-performance computing technology, including LonScript, a programming language for scientific and technical computing, alongside patented data-infrastructure work aimed at reducing the cost of AI computation. The company is raising via equity crowdfunding on Wefunder.
    url: https://www.riecomp.org
    foundingDate: 2018-02-01T00:00:00.000Z
    industry: High-performance computing
    sameAs:
      - https://www.linkedin.com/company/riemann-computing
---

Riemann Computing has completed development of LonScript, a programming language for scientific and technical computing that the Troy, Michigan company says runs 1,434 times faster than Python. The figure is a geometric mean across 15 separate benchmark tests rather than a single result, founder Andrew Kamal says, and the company has published demonstration videos showing the language running sorting, statistics, and analysis workloads, including a quantum circuit simulation.

The language grew out of Kamal's own frustrations with the standard tools of numerical work. "After more than a decade working with tools like NumPy, Wolfram, and MATLAB, I saw where each one hit a wall, whether it was speed, structure, or usability," he says. "LonScript is what happens when you try to fix all three at once."

## A Scientific Computing Language Built for Parallelization

![Andrew Kamal, founder of Riemann Computing](https://cdn.nanimediahouse.com/andrew-kamal-riemann-computing.webp)
*Andrew Kamal, founder of Riemann Computing — Photo: Riemann Computing Inc.*

LonScript targets the same users as Wolfram and MATLAB: mathematicians and physicists running structural analysis, simulations, and heavy numerical work. The difference, Kamal says, is that LonScript is hardware-oriented from the ground up, with a core focus on parallelization and distributed computing rather than single-threaded execution wrapped in a friendlier syntax.

That focus is aimed squarely at artificial intelligence workloads. Training large models is expensive mostly because of the compute hours involved, and a language that shortens numerical operations shortens the bill. "LonScript marks a paradigm shift in computational problem-solving," Kamal says. "Its capacity to execute high-performance tasks with minimal latency moves us closer to safe AGI, making the development of cutting-edge AI models more accessible and scalable."

## The 1,434x Benchmark and How the Company Measured It

The headline number comes from Riemann Computing's own test suite. Across 15 benchmarks, the company reports a geometric mean of 1,434x over Python, with a peak of more than 16,000x on one test, and the company says it also outpaces compiled languages: 23x over C, 22x over C++, and 25x over Julia. The demonstrations are published on the company's YouTube channel, where LonScript runs example programs including the quantum circuit simulation.

The benchmark suite and the language itself have not yet been released for outside testing; Riemann Computing is registering LonScript with the U.S. Copyright Office rather than publishing the code. Until a public toolchain is available, the performance figures rest on the company's own measurements.

## An Equity Crowdfunding Raise on Wefunder

Riemann Computing began life in 2018 as Stark Drones Corporation before taking its current name, and Kamal has kept it small and founder-led. The company is now raising via equity crowdfunding on Wefunder, where the campaign runs up to $1.25 million and pitches LonScript alongside its patented data-infrastructure work as an answer to AI's computing costs. The campaign materials and financials are available to investors on the platform directly.

Kamal is not positioning LonScript as a Python replacement for general software work. The pitch is aimed at the researcher whose simulations run overnight: a language built around parallelization and distributed computing that promises the same work in a fraction of the time.

**About Riemann Computing**

Riemann Computing Inc. is a Troy, Michigan technology company founded by Andrew Kamal in 2018, originally as Stark Drones Corporation. It develops high-performance computing technology, including LonScript, a programming language for scientific and technical computing, alongside patented data-infrastructure work aimed at reducing the cost of AI computation. The company is raising via equity crowdfunding on Wefunder.

[Website](https://www.riecomp.org)

## FAQ

**Q: Which programming language is best for distributed computing?**
There is no single answer; Erlang, Go, and Julia all have strong distributed computing stories, and most large-scale AI work distributes Python across clusters with frameworks doing the heavy lifting. LonScript's approach, the company says, is to build distribution and parallelization into the language's core design rather than into add-on frameworks.

**Q: Is C++ or Python faster?**
Compiled C++ typically runs numerical loops orders of magnitude faster than interpreted Python, which is why Python's scientific stack hands the hard work to compiled libraries like NumPy. Riemann Computing goes further, saying LonScript outpaces even C and C++ in its own 15-test benchmark suite.

**Q: What is a distributed programming language?**
A distributed programming language treats computation spread across many processors or machines as a first-class feature rather than something added through libraries. LonScript is designed on that principle, with parallelization and distributed computing built into its core for scientific workloads.
