Job Overview
Location
San Francisco, CA (Hybrid) OR Remote (Americas, UTC-3 to UTC-10)
Job Type
Full-time
Category
Machine Learning Engineer
Date Posted
March 21, 2026
Full Job Description
đź“‹ Description
- • As a Research Engineer focused on Reinforcement Learning at Firecrawl Inc., you will directly impact the core product by building the training infrastructure, reward pipelines, and fine-tuning systems that enhance the company’s ability to extract, understand, and structure web data at scale. Your work will bridge theoretical RL concepts with practical LLM agent applications, turning research into production-grade improvements that serve real users.
- • Day to day, you will design and operate end-to-end training loops from data collection to deployment, fine-tune foundation models to achieve state-of-the-art performance in web data extraction, and develop reward signals that improve multi-step LLM agent behaviors. You will run rapid experiments to test hypotheses, debug training and convergence issues in GPU environments, and collaborate with the research team to align RL advancements with search, ranking, and product goals.
- • You will join a small, fast-moving, highly technical team at Firecrawl, a company that has achieved 8 figures in ARR and over 90k GitHub stars by delivering the fastest way for developers to get LLM-ready web data. The team values speed, ownership, and shipping impactful work, operating with a strong bias toward iteration and real-world results over academic perfection.
- • In this role, you will deepen your expertise in applying RL to real-world LLM systems, gain end-to-end ownership of model training infrastructure, and learn how to translate complex technical work into clear insights for cross-functional stakeholders. You will have the opportunity to ship models that serve production traffic and directly influence product capabilities in a high-urgency, high-impact environment.
🎯 Requirements
- • 3+ years of experience in applied reinforcement learning, machine learning engineering, or model training with a proven track record of building and deploying models to production systems
- • Demonstrated ability to build training infrastructure, reward pipelines, data pipelines, and evaluation frameworks from scratch without relying on external ML platform teams
- • Fluency in both classical RL methods (e.g., PPO, RLHF, reward modeling) and modern LLM agent systems, with experience bridging the two to improve agent workflows
🏖️ Benefits
- • Competitive salary range of $180,000–$270,000 per year, adjusted fairly for non-U.S. based employees based on local cost of living
- • Equity grant of up to 0.15% in the company, allowing you to own a meaningful stake in what you help build
- • Comprehensive benefits including 100% employer-paid medical, dental, and vision for U.S.-based employees, 12 weeks fully paid parental leave, $100/month wellness stipend, and $1,000/year learning and development stipend
Skills & Technologies
About Firecrawl Inc.
Firecrawl Inc. provides an API that converts entire websites into clean markdown or structured data. Designed for AI applications, the service crawls all accessible subpages, renders dynamic content, and returns LLM-ready output without requiring sitemaps. It includes built-in scraping, search, and extraction capabilities for building knowledge bases, fine-tuning datasets, or powering chatbots. The company targets developers and data teams who need reliable web content ingestion at scale, offering cloud-hosted endpoints and self-hosted options under a usage-based pricing model.
Subscribe to the weekly newsletter for similar remote roles and curated hiring updates.
Newsletter
Weekly remote jobs and featured talent.
No spam. Only curated remote roles and product updates. You can unsubscribe anytime.
Similar Opportunities
Heidi Health Pty Ltd
1 month ago
Heidi Health Pty Ltd
1 month ago
FundraiseUp Inc.
29 days ago