Machine Learning Engineer

distil labs; United States; Remote; USD 180,000 - 240,000; Full time

Benefits offered by this job

Remote work within European time zones

Significant ownership of technical-dec

Production-ready ML systems

Job summary

Distil Labs seeks a senior machine-learning engineer to own the full lifecycle of task-specific language models, from traces and synthetic-data creation through distributed training, evaluation, and production deployment. This role blends hands-on ML engineering with applied research to turn distillation, data generation, and model improvements into repeatable pipelines.

You will build end-to-end workflows, collaborate with researchers to productionize methods, and create benchmarks that measure

Qualifications

• 5+ years building and shipping ML systems in production.
• Ownership of model training, evaluation, and deployment.
• Ability to interpret papers and turn ideas into experiments and pipelines.
• Strong experimental design to distinguish real improvements from noise.
• Experience with Python and modern ML frameworks, distributed training, and GPU infrastructure.
• Comfort working across research, engineering and production ops.

Responsibilities

• Build end-to-end workflows for trace processing, synthetic-data generation, validation, fine-tuning, evaluation, and deployment.
• Partner with research colleagues to test new methods and productionize approaches that improve customer models.
• Create benchmarking and evaluation systems that measure quality, latency, and meaningful model improvement.
• Run and optimize distributed training across cloud and on-premises GPU environments.
• Operate workflow orchestration, container platforms, and secure multi-tenant model-serving systems.
• Help maintain low-latency services that remain reliable under load.

Skills

ML systems in production

Python

Distributed training

Experiment design

Model evaluation & deployment

Tools

PyTorch

TensorFlow

Docker

Kubernetes

Job description

Role overview

Own the full lifecycle of task-specific language models, from customer traces and synthetic-data creation through distributed training, evaluation, and production deployment. This senior role combines hands-on machine-learning engineering with applied research, turning advances in distillation, data generation, and model improvement into repeatable pipelines.

Responsibilities

• Build end-to-end workflows for trace processing, synthetic-data generation, validation, fine-tuning, evaluation, and deployment.
• Partner with research colleagues to test new methods and productionize approaches that improve customer models.
• Create benchmarking and evaluation systems that measure quality, latency, and meaningful model improvement.
• Run and optimize distributed training across cloud and on-premises GPU environments.
• Operate workflow orchestration, container platforms, and secure multi-tenant model-serving systems.
• Help maintain low-latency services that remain reliable under load.

Requirements

• At least five years of experience building and shipping machine-learning systems in production.
• Practical ownership of model training, evaluation, and deployment.
• Ability to interpret research papers and convert ideas into working experiments and pipelines.
• Strong experimental discipline, including designing evaluations that distinguish real improvements from noise.
• Experience with Python and modern machine-learning frameworks, distributed training, and GPU infrastructure.
• Comfort working across research, engineering, and production operations.

Nice to have

• Experience with knowledge distillation, synthetic data, model self-improvement, or small language models.
• Research publications or contributions to open-source machine-learning infrastructure.

Benefits and work setup

• Remote work within European time zones, with periodic in-person team offsites.
• Significant ownership of technical decisions, model quality, and production outcomes.