Prime Intellect

www.primeintellect.ai

1 Job

25 Employees

About the Company

Prime Intellect democratizes AI development at scale. Our platform makes it easy to find global compute resources and train state-of-the-art models through distributed training across clusters. Collectively own the resulting open AI innovations, from language models to scientific breakthroughs.

Listed Jobs

Company Name: Prime Intellect
Job Title: Applied Research - RL & Agents
Job Description: **Job Title:** Applied Research - RL & Agents **Role Summary** Lead applied research and development of advanced AI agents and reinforcement learning (RL) systems to enable real-world deployment of large models. Focus on customer-facing engineering, infrastructure design, and alignment of models with domain-specific tasks through scalable, distributed systems. **Expectations** - Translate customer workflows and requirements into technical solutions for agent systems. - Design, prototype, and deploy next-generation AI agents for automation, decision-making, and reasoning. - Develop and implement RL/post-training methods (e.g., RLHF, RLVR, GRPO) for model alignment. - Architect distributed training/inference pipelines and observability frameworks. - Collaborate with research and product teams to align innovation with practical deployment. **Key Responsibilities** - **Customer Solutions:** Engage directly with customers to understand and resolve workflow challenges using AI agents. - **Agent Development:** Prototype and optimize agentic systems (multi-agent, memory-augmented) for real-world applications. - **Infrastructure Engineering:** Build scalable, performant systems for training, evaluation, and deployment (Kubernetes, Docker, Ray). - **Evaluation Systems:** Create verifiable test harnesses and metrics for reasoning, robustness, and alignment in deployed models. - **Research Integration:** Translate customer needs into technical requirements and research priorities. **Required Skills** - Proficiency in reinforcement learning, post-training, and alignment of large-scale models. - Expertise in distributed systems frameworks (vLLM, sglang, Ray, Accelerate, Kubernetes). - Experience with containerization, orchestration tools (Docker, Terraform), and production pipeline architecture. - Demonstrated research contributions via publications, open-source projects, or benchmarks. - Strong software engineering skills in Python, with a focus on agent systems and RL implementation. **Required Education & Certifications** - Advanced degree (PhD or Master’s) in Machine Learning, Artificial Intelligence, or related field with 3+ years of applied research/engineering experience. - Proven track record in developing production-grade ML/RL systems for real-world use cases.

San francisco, United states

On site

26-09-2025