... RLHF/RLAIF, online RL, and model-based data improvement. • Design the systems abstractions that connect research ideas to production-scale RL runs: trainers, rollout workers, reward models ...
... RLHF/RLAIF, online RL, and model-based data improvement. • Design the systems abstractions that connect research ideas to production-scale RL runs: trainers, rollout workers, reward models ...
Member of Technical Staff - RL Research (Experienced)
$300K - $500K/yr
Develop and scale post-training methods such as PPO, GRPO, DPO, rejection sampling, RLHF/RLAIF, online RL, and model-based data improvement. * Design the systems abstractions that connect research ...
Member of Technical Staff - RL Research (Experienced)
$300K - $500K/yr
Develop and scale post-training methods such as PPO, GRPO, DPO, rejection sampling, RLHF/RLAIF, online RL, and model-based data improvement. * Design the systems abstractions that connect research ...
Experience with agent evaluation, offline/online experiments, and human feedback loops in production. * Direct experience with RLHF, RLAIF, DPO, PPO, GRPO, or related optimization techniques. * Prior ...
Experience with agent evaluation, offline/online experiments, and human feedback loops in production. * Direct experience with RLHF, RLAIF, DPO, PPO, GRPO, or related optimization techniques. * Prior ...
... RLHF/RLAIF, online RL, and model-based data improvement. • Design the systems abstractions that connect research ideas to production-scale RL runs: trainers, rollout workers, reward models ...
... RLHF/RLAIF, online RL, and model-based data improvement. • Design the systems abstractions that connect research ideas to production-scale RL runs: trainers, rollout workers, reward models ...
Develop and scale post-training methods such as PPO, GRPO, DPO, rejection sampling, RLHF/RLAIF, online RL, and model-based data improvement. * Design the systems abstractions that connect research ...
Develop and scale post-training methods such as PPO, GRPO, DPO, rejection sampling, RLHF/RLAIF, online RL, and model-based data improvement. * Design the systems abstractions that connect research ...
Experience with agent evaluation, offline/online experiments, and human feedback loops in production. * Direct experience with RLHF, RLAIF, DPO, PPO, GRPO, or related optimization techniques. * Prior ...
Experience with agent evaluation, offline/online experiments, and human feedback loops in production. * Direct experience with RLHF, RLAIF, DPO, PPO, GRPO, or related optimization techniques. * Prior ...
Build novel online & offline evaluation metrics and methodologies for multimodal personal digital assistants. * Fine-tune/post-train LLMs using techniques like SFT, DPO, RLHF, and RLAIF. * Set up ...
Build novel online & offline evaluation metrics and methodologies for multimodal personal digital assistants. * Fine-tune/post-train LLMs using techniques like SFT, DPO, RLHF, and RLAIF. * Set up ...
Build novel online & offline evaluation metrics and methodologies for multimodal personal digital assistants. * Fine-tune/post-train LLMs using techniques like SFT, DPO, RLHF, and RLAIF. * Set up ...
Build novel online & offline evaluation metrics and methodologies for multimodal personal digital assistants. * Fine-tune/post-train LLMs using techniques like SFT, DPO, RLHF, and RLAIF. * Set up ...
Build novel online & offline evaluation metrics and methodologies for multimodal personal digital assistants. * Fine-tune/post-train LLMs using techniques like SFT, DPO, RLHF, and RLAIF. * Set up ...
Build novel online & offline evaluation metrics and methodologies for multimodal personal digital assistants. * Fine-tune/post-train LLMs using techniques like SFT, DPO, RLHF, and RLAIF. * Set up ...
Build novel online & offline evaluation metrics and methodologies for multimodal personal digital assistants. * Fine-tune/post-train LLMs using techniques like SFT, DPO, RLHF, and RLAIF. * Set up ...
Build novel online & offline evaluation metrics and methodologies for multimodal personal digital assistants. * Fine-tune/post-train LLMs using techniques like SFT, DPO, RLHF, and RLAIF. * Set up ...
Develop and scale post-training methods such as PPO, GRPO, DPO, rejection sampling, RLHF/RLAIF, online RL, and model-based data improvement. * Design the systems abstractions that connect research ...
Develop and scale post-training methods such as PPO, GRPO, DPO, rejection sampling, RLHF/RLAIF, online RL, and model-based data improvement. * Design the systems abstractions that connect research ...
Apply Reinforcement Learning (RLVR, RLHF), Direct Preference Optimization (DPO), and customer ... well as online experiments. About the team Core Search builds the next-generation LLM-powered ...
Apply Reinforcement Learning (RLVR, RLHF), Direct Preference Optimization (DPO), and customer ... well as online experiments. About the team Core Search builds the next-generation LLM-powered ...
Apply Reinforcement Learning (RLVR, RLHF), Direct Preference Optimization (DPO), and customer ... well as online experiments. About the team Core Search builds the next-generation LLM-powered ...
Apply Reinforcement Learning (RLVR, RLHF), Direct Preference Optimization (DPO), and customer ... well as online experiments. About the team Core Search builds the next-generation LLM-powered ...
Build novel online & offline evaluation metrics and methodologies for multimodal personal digital assistants. * Fine-tune/post-train LLMs using techniques like SFT, DPO, RLHF, and RLAIF. * Set up ...
Build novel online & offline evaluation metrics and methodologies for multimodal personal digital assistants. * Fine-tune/post-train LLMs using techniques like SFT, DPO, RLHF, and RLAIF. * Set up ...
Build novel online & offline evaluation metrics and methodologies for multimodal personal digital assistants. * Fine-tune/post-train LLMs using techniques like SFT, DPO, RLHF, and RLAIF. * Set up ...
Build novel online & offline evaluation metrics and methodologies for multimodal personal digital assistants. * Fine-tune/post-train LLMs using techniques like SFT, DPO, RLHF, and RLAIF. * Set up ...
Software Engineer - Global E-Commerce AI Search Infrastructure (TikTok Shop)
Seattle, WA · On-site
$148K - $300K/yr
... latency online services. - Proficiency in C++, Go, or Java (C++ preferred); strong systems ... SFT, RL (RLHF / PPO / GRPO), distillation, or reward model design. - Research publications at ...
Software Engineer - Global E-Commerce AI Search Infrastructure (TikTok Shop)
Seattle, WA · On-site
$148K - $300K/yr
... latency online services. - Proficiency in C++, Go, or Java (C++ preferred); strong systems ... SFT, RL (RLHF / PPO / GRPO), distillation, or reward model design. - Research publications at ...
Senior Applied Scientist
Seattle, WA · On-site
At Amazon Selection and Catalog Systems (ASCS), our mission is to power the online buying ... tuning, RLHF, or agentic architectures Amazon is an equal opportunity employer and does not ...
Senior Applied Scientist
Seattle, WA · On-site
At Amazon Selection and Catalog Systems (ASCS), our mission is to power the online buying ... tuning, RLHF, or agentic architectures Amazon is an equal opportunity employer and does not ...
Senior Software Engineer - Global E-Commerce AI Search Infrastructure (TikTok Shop)
Seattle, WA · On-site
$202K - $368K/yr
... latency online services. - Proficiency in C++, Go, or Java (C++ preferred); strong systems ... SFT, RL (RLHF / PPO / GRPO), distillation, or reward model design. - Research publications at ...
Senior Software Engineer - Global E-Commerce AI Search Infrastructure (TikTok Shop)
Seattle, WA · On-site
$202K - $368K/yr
... latency online services. - Proficiency in C++, Go, or Java (C++ preferred); strong systems ... SFT, RL (RLHF / PPO / GRPO), distillation, or reward model design. - Research publications at ...
Apply Reinforcement Learning (RLVR, RLHF), Direct Preference Optimization (DPO), and customer ... well as online experiments. About the team Core Search builds the next-generation LLM-powered ...
Apply Reinforcement Learning (RLVR, RLHF), Direct Preference Optimization (DPO), and customer ... well as online experiments. About the team Core Search builds the next-generation LLM-powered ...
Applied Scientist
Seattle, WA · On-site
At Amazon Selection and Catalog Systems (ASCS), our mission is to power the online buying ... tuning, RLHF, prompt engineering, or agentic architectures - Experience with LLM/VLM serving ...
Applied Scientist
Seattle, WA · On-site
At Amazon Selection and Catalog Systems (ASCS), our mission is to power the online buying ... tuning, RLHF, prompt engineering, or agentic architectures - Experience with LLM/VLM serving ...
Online Rlhf information
See Seattle, WA salary details
$19.9K - $27K
12% of jobs
$32.1K is the 25th percentile. Wages below this are outliers.
$27K - $34.1K
18% of jobs
$34.1K - $41.2K
15% of jobs
The median wage is $42.2K / yr.
$41.2K - $48.3K
33% of jobs
$48.3K - $55.3K
12% of jobs
$55.3K - $62.4K
0% of jobs
$62.4K - $69.5K
0% of jobs
$69.5K - $76.6K
0% of jobs
$76.6K - $83.7K
0% of jobs
$83.7K - $90.8K
0% of jobs
$90.8K - $97.9K
9% of jobs
$19.9K
$46.2K
$97.9K
How much do online rlhf jobs pay per year?
As of Aug 8, 2026, the average yearly pay for online rlhf in Seattle, WA is $46,199.00, according to ZipRecruiter salary data. Most workers in this role earn between $28,400.00 and $49,500.00 per year, depending on experience, location, and employer.
What are some common challenges faced by online RLHF specialists when collaborating with cross-functional teams?
Online RLHF specialists often work closely with machine learning engineers, data annotators, and product managers. A common challenge is ensuring that feedback from human annotators is accurately integrated into model training, which requires clear communication and well-defined annotation guidelines. Additionally, balancing the pace of model updates with the need for high-quality human feedback can be demanding. Effective collaboration and regular syncs are essential to maintain alignment and achieve project goals.
What is the difference between Online Rlhf vs Online Rlhf?
| Aspect | Online Rlhf | Online Rlhf |
|---|---|---|
| Credentials | Typically requires certification in online health coaching or related fields | Typically requires certification in online health coaching or related fields |
| Work Environment | Remote, online platform-based | Remote, online platform-based |
| Industry Usage | Common in health and wellness sectors | Common in health and wellness sectors |
| Job Focus | Providing health guidance and support online | Providing health guidance and support online |
Online Rlhf and Online Rlhf are the same role, often used interchangeably. Both involve providing health and wellness support remotely, requiring similar certifications and working within the online health industry. The key difference is often in terminology rather than job function.
What is an online RLHF?
Online RLHF (Reinforcement Learning from Human Feedback) jobs typically involve helping to train AI models by providing human feedback on their outputs. Workers in these roles might review model responses, rate the quality of generated text, or suggest improvements to help the AI learn to produce better results. These jobs are often remote and can be done part-time or as contract work. They play a crucial role in improving the safety, usefulness, and accuracy of AI systems by aligning them more closely with human preferences.
What are the key skills and qualifications needed to thrive as an online RLHF specialist, and why are they important?
To thrive as an Online RLHF Specialist, you need a strong background in machine learning, reinforcement learning, and data analysis, typically supported by a degree in computer science or a related field. Familiarity with technical tools like Python, PyTorch or TensorFlow, and experience with human feedback systems or annotation platforms are highly valuable. Strong problem-solving, attention to detail, and the ability to communicate complex concepts clearly are crucial soft skills. These qualifications ensure the effective training and evaluation of AI models, leading to more accurate and reliable machine learning systems.
What are the most commonly searched types of Rlhf jobs in Seattle, WA? The most popular types of Rlhf jobs in Seattle, WA are:
What are popular job titles related to Online Rlhf jobs in Seattle, WA? For Online Rlhf jobs in Seattle, WA, the most frequently searched job titles are:
What job categories do people searching Online Rlhf jobs in Seattle, WA look for? The top searched job categories for Online Rlhf jobs in Seattle, WA are:
What cities near Seattle, WA are hiring for Online Rlhf jobs? Cities near Seattle, WA with the most Online Rlhf job openings:
Full-time
Re-posted 3 days ago
Job description
Job Summary:
Nuance Labs is building innovative AI avatars with emotional intelligence, and they are seeking a Member of Technical Staff to lead reinforcement learning and post-training for large-scale models. The role involves developing and optimizing systems for RL methods and ensuring the scalability and reliability of training systems to enhance interactive behavior and performance.
Responsibilities:
• Build Nuance’s RL/post-training stack from 0→1: rollout generation, policy optimization, reward/reference model serving, data feedback loops, evaluation, checkpointing, observability, and debugging.
• Develop and scale post-training methods such as PPO, GRPO, DPO, rejection sampling, RLHF/RLAIF, online RL, and model-based data improvement.
• Design the systems abstractions that connect research ideas to production-scale RL runs: trainers, rollout workers, reward models, evaluators, data queues, experience buffers, and checkpoint promotion.
• Build evaluation and feedback loops for omni behavior: turn-taking, interruption, timing, emotional response, audiovisual coherence, instruction following, and real-time interaction quality.
• Optimize the end-to-end post-training loop across rollout throughput, serving latency, GPU utilization, policy update efficiency, queueing, checkpoint overhead, and research iteration speed.
• Evolve the platform as algorithms, model architectures, reward definitions, data sources, and evaluation methods change.
Qualifications:
Required:
• Significant hands-on experience with RL, RLHF, RLAIF, post-training, alignment, or large-scale fine-tuning for modern foundation models.
• Deep understanding of RL/post-training methods: policy optimization, reward modeling, preference optimization, rejection sampling, KL control, evaluation, and data feedback loops.
• A track record reasoning about model behavior and training dynamics: reward hacking, unstable rewards, distribution shift, stale policies, mode collapse, over-optimization, noisy preferences, and evaluation mismatch.
• Proven experience building or operating RL/post-training pipelines at scale with frameworks such as verl, ms-swift, OpenRLHF, or equivalent internal systems, including integration with rollout serving systems such as vLLM.
• Experience with large-scale training or inference systems, including rollout generation, model serving, batching, queueing, GPU utilization, checkpointing, and debugging.
• Understanding of omni post-training for real-time audio-video-language interaction: temporal alignment, interruption, emotional response, and multimodal evaluation.
• Strong software engineering fundamentals, curiosity, and adaptability to new RL algorithms, model architectures, serving systems, evaluation methods, and research ideas.
Preferred:
• Prior 0→1 experience building post-training systems, RL pipelines, agent training systems, evaluation platforms, or large-scale model improvement loops.
• Experience with PPO, GRPO, DPO, online RL, RLHF/RLAIF, reward modeling, preference data, synthetic data generation, or model-based data improvement.
• Experience with omni or multimodal post-training for audio-video-language models, especially long-context or real-time interactive systems.
• Experience scaling mixed training/inference workloads across large GPU clusters.
• Experience with adjacent areas such as distributed pretraining, data infrastructure, inference serving, simulation, human/AI feedback collection, or evaluation infrastructure.
• Publications or substantial open-source contributions in RL, post-training, alignment, evaluation, ML systems, or model behavior.
Company:
Nuance Labs an AI research company is developing the first human foundation model that understands and displays emotion in real time. Founded in 2024, the company is headquartered in Seattle, USA, with a team of 11-50 employees. The company is currently Early Stage.
Nuance Labs is building innovative AI avatars with emotional intelligence, and they are seeking a Member of Technical Staff to lead reinforcement learning and post-training for large-scale models. The role involves developing and optimizing systems for RL methods and ensuring the scalability and reliability of training systems to enhance interactive behavior and performance.
Responsibilities:
• Build Nuance’s RL/post-training stack from 0→1: rollout generation, policy optimization, reward/reference model serving, data feedback loops, evaluation, checkpointing, observability, and debugging.
• Develop and scale post-training methods such as PPO, GRPO, DPO, rejection sampling, RLHF/RLAIF, online RL, and model-based data improvement.
• Design the systems abstractions that connect research ideas to production-scale RL runs: trainers, rollout workers, reward models, evaluators, data queues, experience buffers, and checkpoint promotion.
• Build evaluation and feedback loops for omni behavior: turn-taking, interruption, timing, emotional response, audiovisual coherence, instruction following, and real-time interaction quality.
• Optimize the end-to-end post-training loop across rollout throughput, serving latency, GPU utilization, policy update efficiency, queueing, checkpoint overhead, and research iteration speed.
• Evolve the platform as algorithms, model architectures, reward definitions, data sources, and evaluation methods change.
Qualifications:
Required:
• Significant hands-on experience with RL, RLHF, RLAIF, post-training, alignment, or large-scale fine-tuning for modern foundation models.
• Deep understanding of RL/post-training methods: policy optimization, reward modeling, preference optimization, rejection sampling, KL control, evaluation, and data feedback loops.
• A track record reasoning about model behavior and training dynamics: reward hacking, unstable rewards, distribution shift, stale policies, mode collapse, over-optimization, noisy preferences, and evaluation mismatch.
• Proven experience building or operating RL/post-training pipelines at scale with frameworks such as verl, ms-swift, OpenRLHF, or equivalent internal systems, including integration with rollout serving systems such as vLLM.
• Experience with large-scale training or inference systems, including rollout generation, model serving, batching, queueing, GPU utilization, checkpointing, and debugging.
• Understanding of omni post-training for real-time audio-video-language interaction: temporal alignment, interruption, emotional response, and multimodal evaluation.
• Strong software engineering fundamentals, curiosity, and adaptability to new RL algorithms, model architectures, serving systems, evaluation methods, and research ideas.
Preferred:
• Prior 0→1 experience building post-training systems, RL pipelines, agent training systems, evaluation platforms, or large-scale model improvement loops.
• Experience with PPO, GRPO, DPO, online RL, RLHF/RLAIF, reward modeling, preference data, synthetic data generation, or model-based data improvement.
• Experience with omni or multimodal post-training for audio-video-language models, especially long-context or real-time interactive systems.
• Experience scaling mixed training/inference workloads across large GPU clusters.
• Experience with adjacent areas such as distributed pretraining, data infrastructure, inference serving, simulation, human/AI feedback collection, or evaluation infrastructure.
• Publications or substantial open-source contributions in RL, post-training, alignment, evaluation, ML systems, or model behavior.
Company:
Nuance Labs an AI research company is developing the first human foundation model that understands and displays emotion in real time. Founded in 2024, the company is headquartered in Seattle, USA, with a team of 11-50 employees. The company is currently Early Stage.