Techila Technologies offers the Techila Distributed Computing Engine, a middleware platform that lets data scientists, engineers, and researchers scale MATLAB, Simulink, Python, R, Julia, or .NET workloads from a laptop to tens of thousands of cloud CPUs with minimal code changes. The solution provides language‑specific APIs and IDE integrations that automatically provision, schedule, and execute jobs in the cloud, delivering real‑time results while eliminating the need to manage traditional high‑performance computing clusters.
Funding
Funding not disclosed
Founders
Product
Problem
Researchers and engineers often face long runtimes when running computationally intensive code on local machines, and acquiring or managing traditional high‑performance computing clusters is costly and complex. This limits the speed of data analysis, model training, and simulation tasks across many scientific and engineering domains.
Solution
Techila Distributed Computing Engine (DCE) is a middleware platform that lets users scale their MATLAB, Simulink, Python, R, Julia, or .NET workloads from a laptop or workstation to tens of thousands of cloud CPUs with minimal code changes. Language‑specific APIs are provided as part of the Techila SDK, enabling developers to annotate functions or loops and have the engine automatically provision, schedule, and execute the jobs in the cloud. The system configures capacity in under two minutes, supports spot instances to reduce costs, and streams results back in real time to the user's preferred IDE. No additional licenses for MATLAB/Simulink are required, and container or Conda environments are supported to preserve complex dependencies.
Target Audience
Primary customers are data scientists, engineers, and researchers in academia, R&D, and industry who need to accelerate MATLAB, Simulink, Python, R, or .NET computational workloads without managing traditional HPC resources.
Features
- Language‑specific APIs (e.g., @techila.distributable for Python, cloudfor for MATLAB/R) that require only small code modifications
- Seamless integration with popular IDEs such as Jupyter, Visual Studio Code, PyCharm, RStudio, and MATLAB
- Automatic provisioning of 10,000+ vCPUs in under two minutes, with dynamic scaling up or down during execution
- Support for containerized workloads and Conda environments to ensure reproducible execution across the cloud
- Real‑time streaming of results back to the originating IDE for interactive monitoring
- Spot‑instance utilization without MPI dependencies to lower cloud infrastructure costs
- Production‑ready APIs that enable fully automated, end‑to‑end workflows