You will get JAX code from your PyTorch/Tensorflow/SKLearn code
Rising Talent
Project details
Do you want to take advantage of JAX’s speed, automatic differentiation, and clean NumPy-like syntax? I’ll convert your existing ML or scientific code to fully functional JAX implementations — fast, accurate, and ready to run.
I’m a Physics/Plasma MSc experienced in numerical modeling and differentiable programming. My focus is not just on rewriting syntax, but ensuring your new JAX code matches the behavior, gradients, and results of your original framework.
You’ll receive well-structured, reproducible code with clear comments and setup instructions. Whether you need a small model, full project migration, or performance optimization using Flax/Optax, I ensure correctness and readability.
Perfect for researchers, engineers, and developers exploring JAX for optimization, simulation, or machine learning. Let’s modernize your codebase with speed and precision!
I’m a Physics/Plasma MSc experienced in numerical modeling and differentiable programming. My focus is not just on rewriting syntax, but ensuring your new JAX code matches the behavior, gradients, and results of your original framework.
You’ll receive well-structured, reproducible code with clear comments and setup instructions. Whether you need a small model, full project migration, or performance optimization using Flax/Optax, I ensure correctness and readability.
Perfect for researchers, engineers, and developers exploring JAX for optimization, simulation, or machine learning. Let’s modernize your codebase with speed and precision!
Machine Learning Tools
NumPy, pandas, Python, Python Scikit-Learn, PyTorch, scikit-learnWhat's included
| Service Tiers |
Starter
$20
|
Standard
$40
|
Advanced
$70
|
|---|---|---|---|
| Delivery Time | 2 days | 4 days | 7 days |
Number of Revisions | 1 | 0 | 0 |
Model Validation/Testing | - | - | - |
Model Documentation | - | - | - |
Data Source Connectivity | - | - | - |
Source Code | - | - | - |
About Carlos
Physics Engineer & Data Scientist
Gent, Belgium - 1:58 am local time
My expertise bridges physics and data science, allowing me to build models that are both accurate and physically meaningful.
Here’s what I can help you with:
💨 Surrogate Modeling for Simulations: Create fast, physics-informed ML surrogates to replace slow simulation codes — achieving 10×–100× speed-ups with quantified error.
🔬 Experiment Optimization: Use Bayesian optimization and active learning to minimize experimental cost or maximize yield with fewer runs.
I work with Python (NumPy, Pandas, Scikit-learn, PyTorch, Optuna, Streamlit) and deliver clear, documented, and reproducible code — plus concise reports you can act on.
Let’s discuss your project and design a data-driven solution that saves time, reduces risk, or accelerates your R&D.
Steps for completing your project
After purchasing the project, send requirements so Carlos can start the project.
Delivery time starts when Carlos receives requirements from you.
Carlos works on your project following the steps below.
Revisions may occur after the delivery date.
Code Review
After purchase, you share your PyTorch, TensorFlow, or Scikit-learn code and explain what part needs conversion. I review the files to understand the structure, dependencies, and model logic.
Conversion & Adaptation
I translate your code to JAX, adjusting syntax, tensor operations, and model definitions. If relevant, I integrate Flax or Optax for modern JAX training workflows.