Senior Python Engineer

ABOUT THE EMPLOYER

Our clients are building an AI-powered platform that digitizes, stores, and extracts intelligence from deal jacket documents. Dealers can use their platform to cut audit response time from hours to seconds, surface compliance gaps before regulators do, and ask plain-English questions across thousands of deals and get grounded, cited answers instantly.This is a foundational engineering hire.You’ll own the AI and backend systems that make our client’s Smart AI tier work — the pipeline that ingests and classifies deal jacket documents, the RAG layer that powers natural language queries, and the eval system that tells us whether answers are accurate and trustworthy.You’ll also own the AWS infrastructure your work runs on. They’re a small team, so you’ll have real ownership — and real impact.

WHAT WILL YOU WORK ON?

• Own the end-to-end RAG pipeline — from document chunking and embedding through retrieval and grounded answer generation with page-level citations

• Define retrieval quality metrics and make architectural decisions on chunking strategy, embedding models, and vector search when accuracy drops

• Build and operate the eval harness that measures answer quality and acts as a regression gate in CI before changes reach production

• Own the ingestion pipeline that processes real-world deal jacket documents through OCR, document classification, and structured entity extraction

• Design the extraction schema that turns unstructured dealer paperwork into reliable, queryable data that the platform’s compliance features depend on

• Architect and maintain the Python async services, task queues, and Postgres data layer with the reliability standards that compliance-sensitive workflows require

• Own the AWS infrastructure end-to-end (RDS, S3, KMS, IAM, VPC, ECS/Fargate), including IaC in Terraform or CDK and the CI/CD pipelines that deliver changes safely to production

WHAT SKILLS AND EXPERIENCE ARE WE LOOKING FOR?

• 5+ years of backend engineering experience, predominantly Python

• 1–2 years of production AI/LLM experience with measurable accuracy targets

• Hands-on experience building and debugging RAG pipelines, including retrieval quality tuning

• Experience building eval harnesses that measure AI performance and catch regressions before production

• Background in document extraction — OCR, classification, entity tagging — with exposure to messy, real-world document formats

• Solid AWS experience at production scale: IAM, VPC, RDS, S3, ECS/Fargate — beyond just deploying to an existing setup

• Hands-on Terraform or CDK ownership — written from scratch, not just reviewed

ARE THERE ANY OTHER REQUIREMENTS?

• Experience in a regulated industry where data accuracy and audit trails matter (legal, financial, automotive, healthcare)

12:00-21:00, 5/2, UTC+5