AI Research Engineer - Reinforcement Learning (100% Remote)
Join to apply for the AI Research Engineer - Reinforcement Learning (100% Remote) role at Tether.io
At Tether, we pioneer a global financial revolution, building reserve‑backed tokens that empower businesses worldwide. Our solutions enable instant, secure, and low‑cost digital token transactions, underpinned by transparency and trust.
About the Job
As a member of the AI model team, you will drive innovation in reinforcement learning for advanced models, optimizing decision‑making and adaptive behavior to deliver enhanced intelligence and performance in real‑world challenges.
Responsibilities
- Develop and implement state‑of‑the‑art reinforcement learning algorithms to optimize decision‑making in simulated and real‑world settings.
- Build, run, and monitor controlled reinforcement learning experiments, tracking key performance indicators and comparing outcomes against benchmarks.
- Curate high‑quality simulation environments and training datasets tailored to domain challenges, ensuring resources enhance learning and model performance.
- Debug and optimize the reinforcement learning pipeline, addressing reward signal noise, exploration strategies, and policy divergence to improve convergence and stability.
- Collaborate cross‑functionally to integrate reinforcement learning agents into production systems, defining success metrics such as real‑world performance improvements.
Qualifications
A degree in Computer Science or related field; ideally a PhD in NLP, Machine Learning, or a related area.Proven experience with large‑scale reinforcement learning experiments, including online RL techniques such as Group‑Relative Policy Optimization (GRPO).Deep understanding of reinforcement learning algorithms, including online RL methods and gradient‑based optimization (policy gradients, actor‑critic, GRPO).Strong expertise in PyTorch and reinforcement learning frameworks, with practical experience developing RL pipelines from simulation to production.Demonstrated ability to apply empirical research to overcome RL challenges such as sample inefficiency, exploration‑exploitation tradeoffs, and training instability.Seniority Level
Not Applicable
Employment Type
Full‑time
Job Function
Information Technology
Industries
Technology, Information and Internet
#J-18808-Ljbffr