Machine Learning Engineer
Machine Learning Engineer
$170,000 - $220,000/yearIn addition to a competitive salary, this role includes a significant equity stake in an early-stage, YC-backed startup.
Machine Learning Engineer
In addition to a competitive salary, this role includes a significant equity stake in an early-stage, YC-backed startup.
About the Role
About Proxis
Proxis (YC S24) is on a mission to redefine the economics of artificial intelligence. The power of Large Language Models is undeniable, but their cost is a critical barrier to widespread adoption. We're building the first dedicated platform for LLM distillation and serving, designed to unlock production-ready models at 1/10th the cost. We are a small, deeply technical team of builders obsessed with pushing the boundaries of what's possible in AI efficiency.
The Role
As one of our first Machine Learning Engineers, you won't just be using tools—you'll be building the engine. You will be at the heart of our core technology, designing and implementing the novel distillation and optimization algorithms that power our platform. This isn't about incremental improvements; it's about creating step-change advancements in model performance and efficiency. You will work directly with the founders to architect, build, and scale the systems that make smaller, faster, and cheaper LLMs a reality for developers everywhere.
What You'll Do
- Design, implement, and refine state-of-the-art knowledge distillation pipelines for massive language models.
- Develop and apply advanced quantization and pruning techniques to compress models while preserving critical capabilities.
- Build robust benchmarking systems to rigorously evaluate model performance, latency, and cost across various hardware.
- Collaborate with the founding team to shape the product roadmap and translate cutting-edge research into a scalable, commercial platform.
- Own the end-to-end lifecycle of our core ML models, from research and experimentation to production deployment and monitoring.
Who You Are
- A first-principles thinker who is passionate about the low-level details of how models work.
- A builder at heart, driven by the desire to create tangible, high-impact technology.
- Obsessed with performance and efficiency; you find joy in shaving off milliseconds and megabytes.
- Comfortable with ambiguity and excited by the challenge of solving unsolved problems.
- You thrive in a fast-paced environment and want to be a foundational member of a category-defining company.
Must Haves
Essential requirements for this position
- Proven experience in training and fine-tuning Large Language Models (LLMs).
- Deep understanding and hands-on experience with model optimization techniques such as knowledge distillation, quantization, and pruning.
- Expert-level proficiency in Python and deep learning frameworks, particularly PyTorch.
- Strong foundation in Transformer architectures and their variants.
- Experience deploying machine learning models into production environments with a focus on low-latency and high-throughput serving.
Nice to Haves
Preferred qualifications and extras
- Experience with high-performance computing (HPC) and GPU programming (CUDA, Triton).
- Familiarity with modern MLOps stacks (e.g., Kubernetes, Docker, MLflow, Kubeflow).
- Contributions to major open-source ML/AI projects.
- Published research in top-tier AI/ML conferences (NeurIPS, ICML, ICLR, etc.).
- Experience building ML infrastructure from the ground up.
Relevant Skills
Equal Opportunity
Proxis is an equal opportunity employer committed to diversity and inclusivity. We never discriminate based on race, national origin, gender, gender identity, sexual orientation, protected veteran status, disability, age, or other legally protected status. If you require reasonable accommodations due to religious beliefs, pregnancy, or disabilities, let us know at any time.