Work on advanced TensorFlow engineering tasks involving framework internals, custom C++/CUDA operations, distributed training, graph optimization, and runtime performance for cutting-edge AI research projects.
JAX Internals Expert
Remote (Global)
$75 – $75/hr
July 16, 2026
Job description
Role Overview
Role Overview
AfterQuery is seeking experienced JAX Internals Experts with deep knowledge of the framework's compiler, transformation, distributed execution, and kernel internals. This role is designed for machine learning engineers, systems engineers, and researchers who have worked directly with advanced JAX internals to build high-performance ML systems.
This is a fully remote contract opportunity with rolling onboarding, offering flexible work on cutting-edge AI and compiler engineering projects.
Key Responsibilities
Complete Advanced JAX Tasks
- Complete assigned technical tasks involving JAX.
- Develop, optimize, and troubleshoot advanced JAX implementations.
- Ensure code executes correctly throughout the full runtime, including long-running unattended execution.
Work with JAX Internals
- Build or maintain functionality involving:
- Custom transformations
- Custom primitives
- Lowering rules
- XLA compiler integration
- MLIR compiler infrastructure
- Distributed sharding
- Custom kernel authoring
Optimize Distributed ML Systems
- Work with:
- grad
- vmap
- pmap
- custom_vjp
- custom_jvp
- pjit
- jax.Array
- GSPMD
- Pallas kernels
Deliver High-Quality Implementations
- Complete technical assignments independently.
- Ensure reliability, correctness, scalability, and performance.
- Work within the rolling onboarding and task assignment process.
Required Qualifications
Full-time professional or research experience using JAX.
Demonstrated internals-level expertise in one or more of:
- Custom transformations (grad, vmap, pmap)
- custom_vjp
- custom_jvp
- Custom primitives and lowering rules
- XLA compiler development
- MLIR compiler work
- Distributed sharding (pjit, jax.Array, GSPMD)
- Custom kernel authoring using Pallas
Access to suitable hardware, including:
- GPU-enabled workstation
- TPU-enabled machine
- Cloud GPU/TPU instance
Preferred Qualifications
Background in:
- Machine Learning Engineering
- Systems Engineering
- Distributed Systems Engineering
Contributions to:
- JAX
- XLA
- Flax
- Haiku
- Optax
- Equinox
- Related open-source machine learning projects
Compensation
- $75 USD per hour
Work Arrangement
- Remote
- Independent contractor
- Flexible schedule
- Rolling onboarding
Apply now
You will be redirected to the company's website to complete your application.
Intelligent machines powered by Artificial Intelligence computers that can learn, reason, and interact with people are no longer science fiction. Today, a self-driving vehicle powered by AI can meander through a country road at night and find its way.
Our work at NVIDIA is dedicated towards a computing model focused on visual and AI computing. For two decades, NVIDIA has pioneered visual computing, the art and science of computer graphics, with our invention of the GPU.
Who are we? We are a team of obsessive thinkers, builders, and doers from Harvard, CMU, UChicago, UPenn, GT, and more. We have published papers, own patents, built companies, and received international accolades.