Applied ML Engineer
Core
Build end-to-end systems at the intersection of machine learning research and production software, turning promising research methods into reliable, measurable, and usable products for users.
Role type
Applied ML Engineer (Research-to-Production)
Builds
Production-quality evaluation infrastructure, experiment runners, verification workflows, and product interfaces for model analysis.
Domain
Machine Learning, Model Evaluation, LLM Inference, Software Engineering
Required skills
Python, PyTorch, Hugging Face Transformers, ML evaluation design, reading research papers, production software engineering, API development, React/TypeScript frontend integration
Preferred skills
Model provenance/fingerprinting, activation analysis, DSPy/LiteLLM/Temporal/Ray/vLLM, Next.js, GPU model serving, adversarial evaluation design
Technologies
PyTorch, Hugging Face, React, TypeScript, Next.js, PostgreSQL, pgvector, Ray, vLLM, Temporal, DSPy, LiteLLM
Responsibilities
Reproduce and evaluate research methods using open-weight models; Design evaluation datasets, probes, scoring methods, and experiment harnesses; Build and extend evaluation infrastructure including runners and orchestration; Turn research workflows into product experiences with APIs and dashboards; Investigate verification methods under model modification (fine-tuning, merging, quantization); Ship production-quality systems with observability and testing.
Seniority
Mid-Senior, hands-on IC