Research Engineer

Seldon Technologies; Vevay (IN); On-site; USD 133,000 - 182,000; Full time

Benefits offered by this job

Visa sponsorship

Relocation support

Professional development budget

GPU access

Mac hardware

Meal stipend

Job summary

Seldon Technologies is seeking a Research Engineer to transform experimental code into production-grade pipelines for data generation, training, and evaluation. You will collaborate with researchers to build scalable infrastructure, ensure reproducible experiments, and optimize GPU usage in a cutting-edge AI environment.

Responsibilities include converting prototypes to reusable pipelines, distributed experiment support, profiling performance, and maintaining robust tooling for testing,

Qualifications

• Strong Python proficiency and hands-on experience with PyTorch, JAX, or comparable machine learning frameworks.
• Track record building and debugging software for model training, inference, or data processing at scale.
• Solid understanding of machine learning experiments and how data, numerical precision, and implementation choices impact results.
• Software engineering fundamentals: testing, profiling, version control, and documentation.
• Self-directed approach with follow-through; principled decision-making, comfort with uncertainty, and collaborative problem-solving.
• Experience with distributed training or data processing tools (PyTorch distributed, DeepSpeed, Ray) is a plus.
• Familiarity with Hugging Face Transformers, vLLM, or experiment tracking platforms (Weights & Biases, MLflow) is valuable.
• Background in reinforcement learning systems or processing video, audio, and image datasets is a bonus.

Responsibilities

• Convert research prototypes into tested, reusable data generation, training, and evaluation pipelines.
• Build distributed experiment support, including data loading, checkpointing, and agent interaction collection.
• Profile workloads and optimize GPU utilization, memory efficiency, and data throughput.
• Implement tests, experiment tracking, and debugging tools while ensuring engineering changes preserve experimental integrity.
• Collaborate with researchers to translate ideas into robust, maintainable code.

Skills

Python

PyTorch

JAX

Distributed training

Testing

Profiling

Version control

Documentation

Experiment tracking

Transformers

Weights & Biases

MLflow

Reinforcement learning

Video processing

Audio processing

Job description

About the Role

You'll work alongside researchers at Seldon Global to transform experimental code into production-ready systems. As a Research Engineer, you'll build, test, and scale post-training pipelines for generating, processing, and evaluating research data. You'll shape implementation from the earliest experiments and develop shared software infrastructure that grows with the team's research.

What You'll Do

• Convert research prototypes into tested, reusable data generation, training, and evaluation pipelines.
• Build distributed experiment support, including data loading, checkpointing, and agent interaction collection.
• Profile workloads and optimize GPU utilization, memory efficiency, and data throughput.
• Implement tests, experiment tracking, and debugging tools while ensuring engineering changes preserve experimental integrity.
• Collaborate with researchers to translate ideas into robust, maintainable code.

What We're Looking For

• Strong Python proficiency and hands-on experience with PyTorch, JAX, or comparable machine learning frameworks.
• Track record building and debugging software for model training, inference, or data processing at scale.
• Solid understanding of machine learning experiments and how data, numerical precision, and implementation choices impact results.
• Software engineering fundamentals: testing, profiling, version control, and documentation.
• Self-directed approach with follow-through; principled decision-making, comfort with uncertainty, and collaborative problem-solving.
• Experience with distributed training or data processing tools (PyTorch distributed, DeepSpeed, Ray) is a plus.
• Familiarity with Hugging Face Transformers, vLLM, or experiment tracking platforms (Weights & Biases, MLflow) is valuable.
• Background in reinforcement learning systems or processing video, audio, and image datasets is a bonus.

Compensation & Benefits

• Competitive compensation plus equity.
• Comprehensive health, dental, and vision coverage.
• Visa sponsorship and relocation support.
• Professional development budget, top-tier AI tools, GPUs, Mac hardware, and necessary subscriptions.
• Team off-sites, conference attendance, transportation support, gym and sports club memberships, and food stipend with near-full meal coverage.

Location

Full-time or internship position, on-site in Zurich, Switzerland.