About the Role Mercor is partnering with a leading AI research lab to support a Frontier Code Agents project. This role focuses on evaluating and improving frontier AI coding models through realistic machine learning engineering workflows and structured technical assessments.…
PyTorch Internals Expert — Project Torch
Remote (Global)
$50 – $100/hr
July 3, 2026
Job description
AfterQuery is seeking experienced machine learning, systems, and distributed systems engineers with deep, internals-level PyTorch expertise to complete advanced technical tasks for Project Torch.
This fully remote contract opportunity focuses on low-level PyTorch development, including custom autograd implementations, CUDA and tensor extensions, ATen internals, and distributed training systems. Selected contributors will complete up to three technical assignments that evaluate real-world expertise in PyTorch internals.
Key Responsibilities
Complete Advanced PyTorch Tasks
Work independently on up to three technical assignments involving low-level PyTorch development and runtime optimization.
Develop and Validate Solutions
Ensure all code executes successfully through both active development and extended unattended runtime.
Work with PyTorch Internals
Tasks may involve areas such as:
- Custom autograd functions
- CUDA and tensor extensions
- ATen-level development
- Distributed training internals
- Runtime optimization
- Low-level PyTorch implementation
Ideal Qualifications
- Professional or research experience working full-time with PyTorch.
- Demonstrated expertise with PyTorch internals.
- Ability to work independently on advanced technical assignments.
- Access to GPU-enabled hardware or a cloud GPU environment.
Preferred Qualifications
- Background in machine learning, systems engineering, or distributed systems.
- Experience with distributed training internals.
- Compiler or graph-level optimization experience.
- Numerical or runtime optimization expertise.
- Contributions to PyTorch or related open-source projects.
Compensation
- $65 per completed task (up to 3 tasks)
- $100 bonus for successfully completing all three tasks and passing review
Project Details
- Fully remote
- Contract engagement
- Rolling onboarding
- Most assignments require approximately 2 hours of active work, with additional unattended runtime
About AfterQuery
AfterQuery is an AI research laboratory exploring the frontiers of artificial intelligence through novel datasets and experimentation.
Backed by Y Combinator and Box Group, AfterQuery partners with leading AI laboratories to advance next-generation AI systems.
Apply now
You will be redirected to the company's website to complete your application.
About the Role Mercor partners with frontier AI labs to build the evaluations their coding models are trained and measured against. As a eval author / ML engineer, you'll define what "correct" means for the next generation of coding models. What You'll Do Design ML/LLM…
Senior Machine Learning Engineer - AI Data Trainer Location: Remote About the job At Alignerr, we partner with the world’s leading AI research teams and labs to build and train cutting-edge AI models. This project is about capturing how an LLM reasons and takes actions step…
Machine Learning Engineer - AI Data Trainer Location: Remote About the job At Alignerr, we partner with the world’s leading AI research teams and labs to build and train cutting-edge AI models. This project is about capturing how an LLM reasons and takes actions step by step…