Remote: Research Engineer, Large-Scale Web Crawling

ElevenLabs; Greater London; Hybrid; GBP 80,000 - 120,000; Full time

Benefits offered by this job

Innovative culture

Growth paths

Learning & development

Social travel stipend

Annual offsite

Co-working stipend

Job summary

ElevenLabs is seeking a Research Engineer to join its research team, focused on large-scale web crawling for frontier AI models. You will own crawling systems sourcing world-class data from the open web and thrive on building and operating distributed crawlers that scale to billions of pages.

You will tackle content extraction, deduplication, freshness and recrawl strategies, and politeness handling, while designing targeted pipelines to locate high-value sources, including audio, video, and

Qualifications

• Hands-on experience building and scaling web crawlers or scraping systems.
• Strong engineering skills in distributed systems at scale (e.g., Kubernetes, queue-based architectures, or custom pipelines).
• Ability to evaluate quality, coverage, and compliance of crawled data and build tooling to measure it.

Responsibilities

• Build and operate large-scale, distributed web crawlers that source data from billions of pages.
• Solve crawling challenges: content extraction from messy HTML, deduplication, freshness/recrawl, politeness, rate limiting.
• Design crawling pipelines to find high-value data sources (audio, video, multilingual) and create clean training-ready datasets.
• Create tooling to researchers to request, monitor, and explore crawled web data quickly and reliably.

Skills

Web crawling

Distributed systems

Data quality evaluation

Tools

Kubernetes

Job description

ElevenLabs is seeking a Research Engineer to join its research team, focused on large-scale web crawling for frontier AI models. You will own crawling systems sourcing world-class data from the open web and thrive on building and operating distributed crawlers that scale to billions of pages.

You will tackle content extraction, deduplication, freshness and recrawl strategies, and politeness handling, while designing targeted pipelines to locate high-value sources, including audio, video, and