
1 - 10 employees
Founded 2023
π€ Artificial Intelligence
βοΈ SaaS
π API
Artificial Intelligence β’ SaaS β’ API
Featherless AI is a serverless AI inference and model hosting provider that offers API access to a large and growing catalog of open-weight models (12,200+), enabling developers and businesses to deploy, fine-tune, and run models at scale without managing servers. The company provides flat subscription pricing with unlimited tokens, GPU orchestration, private/anonymous usage (no logs), and options for enterprise self-hosting or scale units for high concurrency. Featherless AI also operates as an AI research lab focused on open-source and post-transformer model research, claiming significant cost and performance improvements for large models and AI agents.
Likely ghost job
π January 23
π Anywhere in the World
β° Full Time
π Senior
π΄ Lead
π§ AI Research Scientist
π» Ghost score 65%
Where this job comes from
Posted on Featherless AI's careers page β, not through a job board
First seen by Remote Rocketship January 23
Still open on Featherless AI's site. Checked 7 hours ago.

1 - 10 employees
Founded 2023
π€ Artificial Intelligence
βοΈ SaaS
π API
Artificial Intelligence β’ SaaS β’ API
Featherless AI is a serverless AI inference and model hosting provider that offers API access to a large and growing catalog of open-weight models (12,200+), enabling developers and businesses to deploy, fine-tune, and run models at scale without managing servers. The company provides flat subscription pricing with unlimited tokens, GPU orchestration, private/anonymous usage (no logs), and options for enterprise self-hosting or scale units for high concurrency. Featherless AI also operates as an AI research lab focused on open-source and post-transformer model research, claiming significant cost and performance improvements for large models and AI agents.
β’ Design and evaluate training optimization techniques for large models (e.g. optimization algorithms, schedulers, normalization, curriculum strategies) β’ Improve training efficiency and stability across long runs and large datasets β’ Research and implement methods such as: β’ Optimizer and scheduler innovations β’ Mixed-precision, low-precision, and memory-efficient training β’ Gradient noise reduction, scaling laws, and convergence analysis β’ Training-time regularization and robustness techniques β’ Run large-scale experiments, analyze results, and translate findings into actionable improvements β’ Author or co-author research papers, technical reports, or blog posts β’ Collaborate closely with infrastructure and inference teams to ensure training decisions translate to real-world performance
β’ Strong background in machine learning research, with emphasis on training dynamics and optimization β’ Experience training large neural networks (LLMs, multimodal models, or large sequence models) β’ Publication experience in ML venues (e.g. NeurIPS, ICML, ICLR, ACL, EMNLP, COLM, arXiv) or equivalent high-quality open research β’ Solid understanding of: β’ Optimization theory and practice β’ Backpropagation, gradient flow, and training stability β’ Distributed and large-batch training β’ Proficiency in Python and modern ML frameworks (PyTorch preferred) β’ Ability to independently design experiments and reason from data β’ Nice to Have β’ Experience with non-standard architectures (e.g. RNN variants, long-context models, hybrid systems) β’ Experience optimizing training on GPUs at scale (FSDP, ZeRO, custom kernels) β’ Contributions to open-source ML or research codebases β’ Comfort operating in fast-moving, ambiguous startup environments
β’ Freedom to pursue and publish novel research β’ Real influence over core model training decisions β’ Direct access to large-scale experiments and real production constraints β’ A small, senior team that values thinking deeply and shipping thoughtfully
Apply NowImprove your chances of getting an interview by checking your resume score before you apply.