About the role
Quartr's Document Extraction team sits at the core of what makes the platform work: taking raw, unstructured financial content — earnings call slide decks, filings, transcripts — and turning it into clean, structured knowledge that 800+ institutional clients and AI companies depend on. As a senior engineer on this team, you own extraction and structuring pipelines end to end. This is a hands-on role with real depth, sitting at the intersection of backend engineering, NLP/LLM tooling, and document AI.
What you'll do
- Design, build, and maintain production-grade Python pipelines that ingest unstructured financial documents and produce accurate, validated structured output.
- Apply NLP and LLM techniques — prompting, fine-tuning, or retrieval-augmented approaches — to solve real extraction and structuring problems at scale.
- Work with OCR, document AI tooling, and PDF parsing to handle the full messiness of real-world financial content across 50M+ first-party documents.
- Model and evolve data schemas that downstream systems and API consumers can rely on.
- Deploy and operate backend services on AWS using Docker, owning reliability and performance in production.
- Reason about data quality end to end — not just whether the code runs, but whether the output is actually correct.
Who you are
You have a solid track record as a backend engineer with Python as your primary language, shipping production systems that hold up under real load. You've worked hands-on with NLP or LLMs — whether that's prompt engineering, fine-tuning, or building RAG pipelines — and you've dealt with the practical challenges of extracting structured information from PDFs or other unstructured formats. You've designed data schemas and built ETL pipelines that produce validated, trustworthy output, and you're comfortable deploying and managing services on AWS with Docker.
Nice to have: Experience building or maintaining scalable backend architectures with a deliberate focus on reliability and long-term maintainability at data-pipeline scale.
You take deep ownership of your work and are comfortable holding yourself accountable for data quality, not just code correctness. You do well in environments where the best argument wins, regardless of who makes it.
Why join
Quartr is a founding data partner for OpenAI's ChatGPT for Financial Services and a partner with Perplexity — the infrastructure you build here is already embedded in how the broader AI ecosystem accesses financial data. The company raised $18M and covers 16,000+ public companies across 65+ markets. Every employee receives equity. The Stockholm office operates on a hybrid model, and the culture is built around ownership, long-termism, and quality over hierarchy.