Member of Technical Staff - Web Crawl Engineer

Reflection is a research lab making intelligence open and accessible for everyone to use, customize, and build on. We build open models that let anyone control their intelligence and help shape the future of AI. Our mission: make intelligence open and accessible to all.

The web is one of the most important sources of information for frontier AI systems. The quality, coverage, freshness, and diversity of web data directly influence model capabilities.

As a member of the Data Team, your mission is to build and operate large-scale web crawling systems that continuously discover, acquire, and process content from across the internet. You will own the infrastructure that powers web-scale data collection, from URL discovery and scheduling to distributed crawling, content extraction, and dataset delivery.

You will work directly with world-class researchers to understand which parts of the web matter most for model performance and build systems that efficiently acquire high-value content at scale.

This role is ideal for engineers who love building distributed systems, optimizing large-scale crawlers, and solving the unique technical challenges of collecting data from the modern web.

Working closely with our pre-training, infrastructure, and data quality teams, you will:

• →

  Build and operate web-scale crawling infrastructure capable of continuously collecting data across billions of URLs

• →

  Design and optimize URL discovery, prioritization, scheduling, and crawl orchestration systems

• →

  Develop distributed crawlers that efficiently acquire content while respecting site constraints and operational requirements

• →

  Build systems for content extraction, rendering, parsing, and normalization across diverse web formats

• →

  Improve crawl coverage, freshness, efficiency, and quality through measurement and experimentation

• →

  Design infrastructure for large-scale recrawling, change detection, and incremental updates

• →

  Develop specialized crawlers for high-value domains, dynamic websites, and difficult-to-access content sources

• →

  Analyze crawl performance and web coverage to identify gaps, inefficiencies, and opportunities for improvement

• →

  Build observability, monitoring, and reliability systems for large-scale crawl operations

• →

  Debug production issues and continuously improve the performance, scalability, and resilience of crawling infrastructure

• Passionate about web-scale systems and the challenges of collecting information from the internet

• Curious about how web data influences model capabilities and willing to iterate based on downstream results

• Comfortable balancing crawl quality, coverage, freshness, and operational efficiency

• Enjoy working at the intersection of distributed systems, data infrastructure, and AI

• Able to collaborate closely with researchers, infrastructure engineers, and data quality teams

• Experience building large-scale web crawling, search indexing, content acquisition, or internet-scale data collection systems

• Strong understanding of crawling architectures, URL frontier management, scheduling, and distributed crawl coordination

• Experience with large-scale distributed systems using technologies such as Ray, Spark, Beam, Flink, or similar frameworks

• Familiarity with content extraction, HTML parsing, browser automation, rendering systems, and modern web technologies

• Experience operating systems that process petabyte-scale datasets

• Strong systems engineering skills, including reliability, observability, performance optimization, and debugging

• Experience designing experiments and using data to improve crawl quality, coverage, and efficiency

• Excellent communication skills and the ability to reason clearly about system tradeoffs and operational constraints

• Experience building search engines, web indexes, or internet-scale crawling platforms

• Familiarity with anti-bot systems, dynamic web content, browser automation, and large-scale extraction pipelines

• Understanding of how web data is used in training and evaluating large language models

• Experience with distributed storage systems, content deduplication, and web-scale dataset management