ML Engineer, Micro1

About the Role

We are actively seeking highly skilled Machine Learning Experts to contribute to an innovative AI training project. This critical role involves model development, training and inference systems, numerical computing, performance optimization, and extensive use of Python. You will be responsible for creating, solving, reviewing, and validating challenging machine-learning engineering tasks, impacting the core functionality and efficiency of advanced AI systems.

Key Responsibilities

• Model Development: Develop and validate machine-learning models, training pipelines, inference systems, and supporting infrastructure.
• Implementation: Implement model components, data pipelines, evaluation systems, and numerical methods.
• Workflow Automation: Build reproducible programmatic workflows utilizing Python and command-line tools.
• Core ML Operations: Work with tensor operations, automatic differentiation, various model architectures, tokenization, batching, and generation techniques.
• Optimization: Construct reproducible training or inference workflows, optimize memory or throughput, and debug numerical or system-level failures.
• Verification: Verify that resulting implementations satisfy objective correctness and performance requirements.

Requirements

• Education: A Master’s degree or PhD in Computer Science, Machine Learning, Artificial Intelligence, Applied Mathematics, Statistics, Engineering, or a closely related quantitative discipline.
• Experience: Strong professional or research experience in machine learning.
• Programming: Practical proficiency with Python.
• ML Frameworks: Meaningful experience with at least two relevant ML frameworks, libraries, or inference tools.
• System Understanding: Strong understanding of model training, evaluation, numerical computation, or inference.
• Debugging: Ability to debug ML systems beyond surface-level API usage.
• Communication: Ability to explain implementation decisions, performance trade-offs, and failure modes clearly.
• Reproducibility: Experience building reproducible technical workflows.

Relevant Tools & Technologies

Candidates should have practical experience with tools such as:

• PyTorch
• JAX
• NumPy and SciPy
• SGLang
• vLLM
• llama.cpp
• Hugging Face Transformers
• Hugging Face Tokenizers
• Equivalent tools demonstrating directly relevant depth will also be considered.

Preferred Qualifications

• Experience at a well-established technology company, AI laboratory, research organization, or other recognized engineering environment is strongly preferred.
• Exceptional open-source or academic experience may also qualify.

Application Process

1. Apply to the role and complete the screening questions.
2. Complete an AI interview of approximately 30 minutes.
3. Complete the hiring manager review.

Compensation Structure

Compensation for this role is output-based. Experts are paid per task that successfully meets the project specifications. The time required to complete each task may vary based on the expert’s experience and workflow efficiency.

Location

Remote, Worldwide