---
title: 'AI Researcher — Training Optimization at Featherless AI'
canonical: 'https://feeny.ai/job/ai-researcher-training-optimization-featherless-ai-world-hyc2csn67p2p'
type: 'job'
last_seen: '2026-09-08'
---

# AI Researcher — Training Optimization at Featherless AI

- **Company:** Featherless AI
- **Location:** World
- **Employment:** full-time
- **Work type:** remote
- **Posted:** 2026-01-23
- **Last confirmed live:** 2026-09-08
- **Apply:** https://jobs.ashbyhq.com/featherlessai/6dc98a0d-9d44-45aa-9435-c2672f8bdd68

## Job description

## ABOUT THE ROLE

We’re looking for an AI Researcher focused on training optimization to help us push the efficiency, stability, and scalability of large-scale model training. You’ll work at the intersection of research and systems, developing novel techniques to reduce training cost, accelerate convergence, and improve model quality—while validating ideas through rigorous experiments and publications.

This role is ideal for someone who enjoys turning research insights into practical training wins, and who has a track record (or strong ambition) of publishing applied ML research.

## WHAT YOU’LL WORK ON

- Design and evaluate training optimization techniques for large models (e.g. optimization algorithms, schedulers, normalization, curriculum strategies)
- Improve training efficiency and stability across long runs and large datasets
- Research and implement methods such as:
- Optimizer and scheduler innovations
- Mixed-precision, low-precision, and memory-efficient training
- Gradient noise reduction, scaling laws, and convergence analysis
- Training-time regularization and robustness techniques
- Run large-scale experiments, analyze results, and translate findings into actionable improvements
- Author or co-author research papers, technical reports, or blog posts
- Collaborate closely with infrastructure and inference teams to ensure training decisions translate to real-world performance

## WHAT WE’RE LOOKING FOR

- Strong background in machine learning research, with emphasis on training dynamics and optimization
- Experience training large neural networks (LLMs, multimodal models, or large sequence models)
- Publication experience in ML venues (e.g. NeurIPS, ICML, ICLR, ACL, EMNLP, COLM, arXiv) or equivalent high-quality open research
- Solid understanding of:
- Optimization theory and practice
- Backpropagation, gradient flow, and training stability
- Distributed and large-batch training
- Proficiency in Python and modern ML frameworks (PyTorch preferred)
- Ability to independently design experiments and reason from data

## NICE TO HAVE

- Experience with non-standard architectures (e.g. RNN variants, long-context models, hybrid systems)
- Experience optimizing training on GPUs at scale (FSDP, ZeRO, custom kernels)
- Contributions to open-source ML or research codebases
- Comfort operating in fast-moving, ambiguous startup environments

## WHY THIS ROLE

- Real influence over core model training decisions
- Freedom to pursue and publish novel research
- Direct access to large-scale experiments and real production constraints
- A small, senior team that values thinking deeply and shipping thoughtfully

## About Featherless AI

## Company Overview
- **One-liner**: Featherless AI provides a serverless platform that offers API access to over 40,000 open-weight AI models from a single endpoint, designed for developers and enterprises.
- **Entity Type**: Private (Series A)
- **Headquarters**: San Francisco, California, United States
- **Founded**: 2023
- **Founders**: Eugene Cheah (CEO, Co-Founder)

## Core Business
- **Primary industry**: Artificial Intelligence Infrastructure / Serverless LLM Hosting
- **Target customers**: B2B, serving developers, AI startups, and enterprises seeking scalable, cost-effective inference for open-source models.
- **Mission or purpose**: To democratize access to all AI models by making them available for serverless inference, eliminating the need for server setup and complex infrastructure management.

## Products & Services
- **Featherless API**: A unified API gateway providing instant access to over 40,000 open-weight models (e.g., DeepSeek, Llama, Mistral, Qwen, RWKV, GLM, Kimi) without setup or hosting. Pricing is flat-rate with unlimited tokens, starting at $25/month for up to 4 concurrent connections and 32K context, scaling to $200/month for higher tiers. Agent-specific plans ($100/month) include sandbox environments and persistent storage. The service emphasizes low latency, dependable uptime, and predictable costs.

## Market Standing
- **Valuation/Market Cap**: Not publicly disclosed.
- **Key Metric**: **Total Funding of $25M** — raised $5M in a Seed round (April 2025, led by Airbus Ventures) and $20M in a Series A round (announced ~May 2026, details still emerging).
- **Notable Investors/Partners**: Airbus Ventures, Kickstart Ventures, Panache Ventures, BMW i Ventures, AMD Ventures, and 11 other investors. The platform is built by researchers contributing to RWKV, a Linux Foundation project.
- **Growth Signals**: The company is on a rapid growth trajectory, with headcount increasing 46.7% year-over-year to 14 employees. The website boasts over 2,100 "stars" for its Discord community. The company has a global presence, operating in 9 countries (including Singapore, Canada, Czechia, UK, Belgium, and Sweden). Website traffic is strong (73,508 monthly visits, growing +19.9% month-over-month), and there are 43 active job postings, a 65.4% quarterly increase in hiring. The Series A announcement signals significant investor confidence.

## Competitive Advantages
- **Extensive Model Library & Zero-Friction Access**: A single API key provides access to the entire Hugging Face trending library, including models up to 229B parameters, with no need to manage infrastructure.
- **Unlimited-Token, Flat-Rate Pricing**: A strong differentiator in the "per-token" pricing era, offering predictable costs suitable for scaling, with tiers from $25 to $200/month.
- **Build for Reliability & Performance**: Architecture designed for real workloads with low latency and dependable uptime, utilizing proprietary GPU orchestration and model load-balancing.
- **Open-Source Roots & Community**: Built by researchers contributing to RWKV (a Linux Foundation project), the company is deeply embedded in the open-source AI ecosystem, which fosters trust and community-driven development.

## Strategic Focus
- **Scaling the Platform & Enterprise Adoption**: The Series A funding will be used to expand AI infrastructure and grow the platform. The company is actively hiring for senior roles like Founding Account Executives and Business Development Reps, signaling a shift toward aggressive go-to-market and enterprise sales.
- **Expanding Global Presence**: With a distributed team across the US, Europe, and Asia, the company is building a global, remote-first workforce.
- **Technical Innovation**: Focused on continuous improvement of inference performance and cost-efficiency through their GPU orchestration system and model load-balancing.

## Why Work Here
- **High-Growth Stage**: As a Series A startup with strong investor backing, this is an opportunity to join a company experiencing rapid scaling, which offers significant career growth and impact potential.
- **Impact & Ownership**: Employees are likely to have high autonomy and a direct impact on the company's trajectory, from building core infrastructure to driving revenue.
- **Remote-First & Global Team**: Based on the distributed headcount across 9 countries (US, Singapore, Canada, UK, Belgium, etc.), the company is clearly remote-first, offering flexibility in where you work. Job postings reflect opportunities in the US and Europe (e.g., Paris, Berlin).
- **Cutting-Edge Technical Challenge**: The core work involves solving complex problems in AI inference, GPU orchestration, and MLOps, making it a compelling place for engineers and researchers passionate about AI infrastructure.
- **Culture & Values**: The company's deep ties to open-source AI communities and its "flat-rate, no-surprises" pricing philosophy likely translate into a transparent, developer-friendly internal culture. The small, highly-skilled team (14 people) suggests a close-knit, high-performing environment.

## Sources
1. [Featherless.ai Website](https://featherless.ai/)
2. [Featherless AI LinkedIn](https://www.linkedin.com/company/feather-serverless-ai)
3. [Featherless AI Docs](https://featherless.ai/docs/overview)
4. [CB Insights Profile](https://www.cbinsights.com/company/recursal-ai)
5. [Featherless AI Jobs](https://jobs.ashbyhq.com/featherlessai)

## Other roles at Featherless AI

- [Founding Account Executive (AI Cloud)](https://feeny.ai/job/founding-account-executive-ai-cloud-featherless-ai-us-krryw5zpmt01) — US &, Canada
- [Founding Business Development Rep (AI Cloud US/CA)](https://feeny.ai/job/founding-business-development-rep-ai-cloud-us-ca-featherless-ai-us-y876z86dw2y3) — US &, Canada
- [Chief of Staff](https://feeny.ai/job/chief-of-staff-featherless-ai-san-francisco-15ydsmveqwsq) — San Francisco, CA
- [Content Marketer](https://feeny.ai/job/content-marketer-featherless-ai-europe-xnsbz1evvkbs) — Europe
- [Business Development Rep (AI Cloud)](https://feeny.ai/job/business-development-rep-ai-cloud-featherless-ai-europe-hd5fymsqsh5h) — Europe
- [AI Researcher – Multilingual Data](https://feeny.ai/job/ai-researcher-multilingual-data-featherless-ai-world-aj24t2jw442p) — World
- [AI Researcher — AI Architecture Research](https://feeny.ai/job/ai-researcher-ai-architecture-research-featherless-ai-world-dprg8203nt10) — World
- [AI Researcher — Distillation](https://feeny.ai/job/ai-researcher-distillation-featherless-ai-world-zxs1tx4mwq1f) — World
- [AI Researcher — Inference Optimization](https://feeny.ai/job/ai-researcher-inference-optimization-featherless-ai-world-5q35t4rfgjve) — World
- [Machine Learning Engineer — AI Architecture Research](https://feeny.ai/job/machine-learning-engineer-ai-architecture-research-featherless-ai-world-88g5fbqva8pd) — World