Senior AI Pipeline Engineer

About the job

The RoleWe are looking for a hands-on Senior AI Pipeline Engineer (Python) who will: • Own and optimize our production extraction pipelines. • Deliver new document-analysis pipelines end-to-end. • Build the foundation for next-generation GenAI features (agentic contract drafting & interpretation). This is a delivery-focused role. You will own the AI kernel (prompts, evaluation logic, structured outputs, validation, model/tool logic), while product engineers orchestrate workflows via stable APIs. What Youll Do 1. Own & Extend Production Pipelines • Maintain and optimize our Python-based extraction pipelines (AWS Lambda + S3 + Docker components). • Ensure stable document processing and downstream triggering. • Improve observability: logging, metrics, alerting, traceability, cost monitoring. • Debug and stabilize real-world failure modes in production. 2. Deliver New Document Pipelines • Design and implement end-to-end pipelines for new document families. • Build evaluation datasets and regression tests. • Prevent silent quality degradation through measurable metrics. 3. LLM-Based Interpretation & Structured Extraction • Improve Q&A and structured extraction using LLMs. • Implement structured outputs, retrieval (RAG where useful), and deterministic validation. • Add robust failure handling (timeouts, retries, fallbacks, safe defaults). 4. GenAI Foundations • Build agentic building blocks in Python behind stable APIs. • Contribute to a contract-generation/editing kernel (planner, drafter, risk checks). • Collaborate with backend/frontend teams for clean integration. 5. Production Readiness • Ensure scalability, cost-efficiency, and security. • Contribute to deployment/versioning/rollback strategies. • Help define operational runbooks. Must-Have Skills • Strong production-grade Python (clean architecture, testing, packaging, APIs). • Experience owning code in production. • AWS serverless (Lambda + S3 required; Step Functions/SQS/CloudWatch a plus). • Docker and containerized services. • Proven experience maintaining/debugging automated pipelines. • Hands-on experience with LLMs (OpenAI/Azure OpenAI/Anthropic or similar): • Structured outputs • Prompt iteration • Retrieval (RAG) • Evaluation approaches Nice to Have • Document AI experience (OCR, layout extraction, noisy PDFs). • Evaluation-driven development (test sets, regression checks, quality metrics). • Experience with cost/latency budgeting. • Familiarity with TypeScript/Node. • Experience integrating REST services. First 90 Days – What Success Looks Like Weeks 1–2 • Fully understand current pipeline architecture. • Stabilize staging/local environments. • Define quality, cost, and latency baselines. • Improve logging and monitoring. Weeks 3–6 • Deliver a new production-ready pipeline for a new document family. • Implement evaluation datasets and regression checks. • Deploy with monitoring and rollback strategy. Weeks 7–12 • Improve Q&A/extraction accuracy measurably. • Deliver a first version of a GenAI contract drafting kernel. • Harden operations (cost controls, retries, fallbacks, documentation).

Requirements

• Python
• AWS Lambda
• S3
• Docker
• LLMs

Preferred Technologies

• Python
• AWS Lambda
• S3
• Docker
• LLMs

About the company

Inhubber is a security-first, AI-powered Contract Lifecycle Management (CLM) platform built for organizations with high compliance and data protection requirements. We combine end-to-end encryption and modern cloud architecture with advanced AI to extract, analyze, and generate contract intelligence. Our platform processes sensitive legal documents for companies worldwide.