Senior RL Research Engineer — Remote, Equity & Impact

Jobgether; Netherlands; On-site; EUR 90,000 - 130,000; Full time

Benefits offered by this job

Competitive salary and equity

Health, dental, and vision benefits

Remote-friendly environment

Flexible vacation policy

Job summary

Jobgether in the Netherlands is building a small, senior engineering team to advance post-training AI agents for voice and text interactions. You will focus on training environments, data pipelines, and evaluations, driving measurable improvements in model performance at scale.

You will collaborate remotely with experienced engineers, tackle challenges in long-horizon interactions, tool use, and reproducible production deployment, delivering robust AI systems for production scale.

Qualifications

• Must have strong Python engineering skills and production-ready systems.
• Experience taking ML models from raw data through experiments to production.
• Data-centric mindset focusing on coverage, quality and leakage.
• Ability to design robust rewards, verifiers and evaluation mechanisms.
• Practical GPU experience and understanding of their limits.
• Judgment on when training is appropriate vs. simpler approaches.
• Ability to collaborate in a remote, distributed team.
• Experience with post-training techniques like fine-tuning or RL.
• Familiarity with open-weight model families and LoRA.
• Experience building tool-using or multi-turn agents.

Responsibilities

• Build training environments, verifiers, and supporting infrastructure for post-training models.
• Own the synthetic data pipeline from data generation through QA and validation.
• Run end-to-end training experiments, analyze results, and identify drivers of improvements.
• Design and maintain evaluations that models must pass before production releases.
• Select and adapt open-weight foundation models for agent and product needs.
• Develop trained behaviors that work across voice and text agents.
• Deploy trained models to production and improve them from real-world feedback.
• Develop approaches to long-horizon interactions and reliable tool use in live conversations.

Skills

Python

Experimentation

Production ML

Data-centric mindset

GPU experience

Reinforcement learning

Fine-tuning

Reward design

Open-weight models

Remote collaboration

Tooling for verifiers

Multi-turn agents

Tools

TR L

OpenRLHF

vLLM

FSDP

LoRA

Qwen

Llama

SGLang

verl

Job description

Jobgether in the Netherlands is building a small, senior engineering team to advance post-training AI agents for voice and text interactions. You will focus on training environments, data pipelines, and evaluations, driving measurable improvements in model performance at scale.

You will collaborate remotely with experienced engineers, tackle challenges in long-horizon interactions, tool use, and reproducible production deployment, delivering robust AI systems for production scale.