SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
NationGraph is building the data infrastructure to make public sector information accessible and actionable for businesses selling to government entities. The company collects and structures procurement data, contracts, and public records from 100,000+ government entities across the US—a differentiated asset no competitor has solved at scale.
As Senior Software Engineer, Data Engineering, you'll own the systems behind NationGraph's core competitive advantage: the data acquisition platform that ingests public information at massive scale. This is production data engineering, not dashboard work. You'll design and build high-volume ingestion pipelines with proper retry logic, dead-letter handling, checkpointing, and resumability. You'll transform messy government documents (PDFs, spreadsheets, scans, web portals) into clean, structured data flowing reliably into the product.
The role emphasizes AI-native architecture: LLM-driven document extraction, classification, and form processing are already in production, and you'll expand these capabilities to be more modular, rigorous, cost-effective, and faster. You'll also design human-in-the-loop systems pairing automation with an operations team, building internal tooling that meets SOC 2 compliance requirements with proper access controls and audit logging.
You'll own problems end-to-end in a small, senior team. Take a goal like "automate records requests through web portals" from research through design doc to production, seeing your work directly impact customer accounts and revenue. Every decision materially changes the company's trajectory.
Required: 5+ years building production backend systems; strong TypeScript/Node in a monorepo environment; deep pipeline engineering (idempotent processing, retries, dead-letter handling, checkpointing); solid PostgreSQL (schema design, migrations, indexing, optimization); document processing at scale (PDF/spreadsheet extraction, deduplication, edge cases); LLM integration (API calls, rate limiting, cost management, prompt engineering); third-party API fluency (webhooks, pagination, rate limits); operational ownership and instrumentation; end-to-end independence.
Nice-to-have: browser automation/scraping (Playwright/Puppeteer); email infrastructure; internal tools for operations teams; Python/Go; workflow orchestration (Airflow); S3/object-storage architectures; OCR and document AI; SOC 2 and compliance-aware engineering; government/public-records domain experience.
Stack: TypeScript/Node monorepo, React frontend, PostgreSQL, S3, Redis, Docker, Kubernetes, AWS, Python, Go. Work closely with CEO/CTO in a zero-bureaucracy, fast-moving environment with great people and diversity of thought.
