{"schemaVersion":"jobsearcher.job.v1","id":"0bcbc5fe79a5d2e7ab33fefd","url":"https://jobsearcher.com/jobs/0bcbc5fe79a5d2e7ab33fefd","canonicalUrl":"https://jobsearcher.com/jobs/0bcbc5fe79a5d2e7ab33fefd","title":"Research Engineer - Inference","description":"Own Your Intelligence\n\nPrime Intellect is building the open superintelligence stack: the infrastructure frontier AI labs build internally, made available to every ambitious AI team.\n\nOur platform, Lab, unifies compute, environments, evaluations, secure sandboxes, high-performance training, and deployment into one full-stack system for post-training at frontier scale - from SFT and RL to tool use, agent workflows, and continuously improving production models. We are building open frontier AI: open-source models trained end to end for long-horizon tasks like autonomous research, and the full-stack platform our own research team uses to build them. The next generation of AI companies, enterprises, and research teams do not just need more GPUs. They need the ability to turn their own workflows, tools, data, and feedback loops into superintelligence they own.\n\nPrime Intellect has raised $150M in total funding from Founders Fund, Radical Ventures, NVIDIA, and exceptional AI, infrastructure, and enterprise operators, including Andrej Karpathy, Dwarkesh Patel, and leaders and founders from Ramp, Perplexity, Harvey, Mercor, Zapier, Datadog, Cognition, OpenAI, Thinking Machines, Together AI, SemiAnalysis, LangChain, Browserbase, Cloudflare, Sierra, Databricks, Airbnb, OpenRouter, Standard Intelligence, Fleet, Core Auto, and more. We are looking for people who want to build at the intersection of frontier research, real infrastructure, and go-to-market for a category that does not fully exist yet.\n\nRole Impact\n\nBuild the inference systems that power Prime Intellect’s RL research, scaling toward 1–3T parameter models on NVL72 clusters. Your work will improve rollout throughput and make frequent model updates practical.\n\nResponsibilities\n\nExtend prime-rl’s vLLM integration to bring new model architectures into RL training, with small-scale validation before larger runs.\n\nImprove rollout throughput for long-horizon agents, tuning concurrency and KV-cache reuse as context lengths grow.\n\nOptimize multi-node MoE inference, including expert parallelism and the tradeoffs of separating prefill from decode.\n\nReduce trainer-to-inference weight-update overhead through sharding-aware transfer and communication overlap, while keeping rollout workers on the intended policy version.\n\nBuild on FP8 and evaluate lower precision. Use trainer/sampler mismatch checks to catch numerical regressions and measure the effect on RL training.\n\nWork with model researchers on architecture choices that improve inference efficiency, balancing model quality against execution cost on the target hardware.\n\nRequirements\n\nExperience building or optimizing GPU inference systems, with an understanding of inference-engine internals.\n\nStrong Python skills and experience with PyTorch or a comparable ML framework.\n\nAbility to profile end-to-end workloads and turn measurements into verified performance improvements.\n\nExperience debugging numerical or integration issues in ML systems.\n\nNice to have\n\nContributions to vLLM or another inference engine.\n\nCUDA or Triton kernel work, particularly for low-precision inference.\n\nBenefits and Perks\n\nCash Compensation Range of $150-350k, including equity incentives, aligning your success with the growth and impact of Prime Intellect.\n\nFlexible work arrangements, with the option to work remotely or in-person at our offices in San Francisco.\n\nVisa sponsorship and relocation assistance for international candidates.\n\nQuarterly team off-sites, hackathons, conferences and learning opportunities.\n\nOpportunity to work with a talented, hard-working and mission-driven team, united by a shared passion for leveraging technology to accelerate science and AI.\n\nIf you're excited about the opportunity to build the foundation for the future of decentralized AI and create a platform that empowers developers and researchers to push the boundaries of what's possible, we'd love to hear from you.","company":"Primeintellect","rawCompany":"primeintellect","city":"Denver","state":"CO","isRemote":false,"isActive":false,"createdAt":"2026-10-19T12:22:10.307Z","occupations":[{"code":"15-1221.00","title":"Computer and Information Research Scientists","slug":"computer-and-information-research-scientists"},{"code":"15-1299.08","title":"Computer Systems Engineers/Architects","slug":"computer-systems-engineers-architects"},{"code":"15-1252.00","title":"Software Developers","slug":"software-developers"}],"industries":[{"code":"541511","title":"Custom Computer Programming Services","slug":"custom-computer-programming-services"},{"code":"541512","title":"Computer Systems Design Services","slug":"computer-systems-design-services"},{"code":"513210","title":"Software Publishers","slug":"software-publishers"}],"jobPosting":{"@context":"https://schema.org","@type":"JobPosting","title":"Research Engineer - Inference","description":"Own Your Intelligence\n\nPrime Intellect is building the open superintelligence stack: the infrastructure frontier AI labs build internally, made available to every ambitious AI team.\n\nOur platform, Lab, unifies compute, environments, evaluations, secure sandboxes, high-performance training, and deployment into one full-stack system for post-training at frontier scale - from SFT and RL to tool use, agent workflows, and continuously improving production models. We are building open frontier AI: open-source models trained end to end for long-horizon tasks like autonomous research, and the full-stack platform our own research team uses to build them. The next generation of AI companies, enterprises, and research teams do not just need more GPUs. They need the ability to turn their own workflows, tools, data, and feedback loops into superintelligence they own.\n\nPrime Intellect has raised $150M in total funding from Founders Fund, Radical Ventures, NVIDIA, and exceptional AI, infrastructure, and enterprise operators, including Andrej Karpathy, Dwarkesh Patel, and leaders and founders from Ramp, Perplexity, Harvey, Mercor, Zapier, Datadog, Cognition, OpenAI, Thinking Machines, Together AI, SemiAnalysis, LangChain, Browserbase, Cloudflare, Sierra, Databricks, Airbnb, OpenRouter, Standard Intelligence, Fleet, Core Auto, and more. We are looking for people who want to build at the intersection of frontier research, real infrastructure, and go-to-market for a category that does not fully exist yet.\n\nRole Impact\n\nBuild the inference systems that power Prime Intellect’s RL research, scaling toward 1–3T parameter models on NVL72 clusters. Your work will improve rollout throughput and make frequent model updates practical.\n\nResponsibilities\n\nExtend prime-rl’s vLLM integration to bring new model architectures into RL training, with small-scale validation before larger runs.\n\nImprove rollout throughput for long-horizon agents, tuning concurrency and KV-cache reuse as context lengths grow.\n\nOptimize multi-node MoE inference, including expert parallelism and the tradeoffs of separating prefill from decode.\n\nReduce trainer-to-inference weight-update overhead through sharding-aware transfer and communication overlap, while keeping rollout workers on the intended policy version.\n\nBuild on FP8 and evaluate lower precision. Use trainer/sampler mismatch checks to catch numerical regressions and measure the effect on RL training.\n\nWork with model researchers on architecture choices that improve inference efficiency, balancing model quality against execution cost on the target hardware.\n\nRequirements\n\nExperience building or optimizing GPU inference systems, with an understanding of inference-engine internals.\n\nStrong Python skills and experience with PyTorch or a comparable ML framework.\n\nAbility to profile end-to-end workloads and turn measurements into verified performance improvements.\n\nExperience debugging numerical or integration issues in ML systems.\n\nNice to have\n\nContributions to vLLM or another inference engine.\n\nCUDA or Triton kernel work, particularly for low-precision inference.\n\nBenefits and Perks\n\nCash Compensation Range of $150-350k, including equity incentives, aligning your success with the growth and impact of Prime Intellect.\n\nFlexible work arrangements, with the option to work remotely or in-person at our offices in San Francisco.\n\nVisa sponsorship and relocation assistance for international candidates.\n\nQuarterly team off-sites, hackathons, conferences and learning opportunities.\n\nOpportunity to work with a talented, hard-working and mission-driven team, united by a shared passion for leveraging technology to accelerate science and AI.\n\nIf you're excited about the opportunity to build the foundation for the future of decentralized AI and create a platform that empowers developers and researchers to push the boundaries of what's possible, we'd love to hear from you.","datePosted":"2026-10-19T12:22:10.307Z","dateModified":"2026-10-19T12:22:10.307Z","hiringOrganization":{"@type":"Organization","name":"Primeintellect","sameAs":"https://jobsearcher.com"},"jobLocation":{"@type":"Place","address":{"@type":"PostalAddress","addressLocality":"Denver","addressRegion":"CO","addressCountry":"US"}},"identifier":{"@type":"PropertyValue","name":"JobSearcher","value":"0bcbc5fe79a5d2e7ab33fefd"},"url":"https://jobsearcher.com/jobs/0bcbc5fe79a5d2e7ab33fefd"}}