Senior AI Engineer, Unstructured AI

Collibra•New York, NY

4d•$204,000 - $255,000•Hybrid

About The Position

Joining Collibra’s Unstructured AI Team. Work at the forefront of context engineering - shaping how AI systems retrieve, structure, and leverage context to deliver accurate, high-quality results at scale. Own end-to-end technical delivery of Unstructured AI systems - from feature prototype to stable production across enterprise environments. Build and scale full-stack systems that ingest, process, and enrich large volumes of unstructured content from distributed enterprise silos (PDFs, contracts, reports, and other document types). Collaborate with the Best: Work closely with xYC Founders to understand complex business challenges and deliver Deasy to solve them. Be part of a dynamic team where ideas flow freely and creativity thrives. Learn and Lead: Stay ahead of the curve by engaging with the latest developments in machine learning and AI. Share knowledge and lead by example to maintain high building standards.

Requirements

Strong proficiency in Python (data processing, API development, and integrations).
Hands-on work with LLM-based and AI-driven enrichment models (e.g., classification, entity extraction, deduplication, PII detection).
Production experience with Spark or comparable big data frameworks — you've tuned and debugged jobs at real scale, not just written ones that worked on sample data.
Experience shipping tested, reviewed production services rather than notebooks — and the discipline to hold that line when a coding agent writes the first draft.
A track record of starting things: a specific example where you took a vaguely-scoped problem, defined the MVP yourself, wrote down your assumptions, and shipped it — without a PM converting it into tickets first.
Solid understanding of data pipelines, microservice architecture, and API design.
Experience ingesting and processing data from third-party enterprise sources (e.g., SharePoint/OneDrive, Salesforce, and SaaS-based knowledge bases).
Knowledge of model evaluation best practices.
Experience with search relevance.
A bachelor’s degree or equivalent related working experience is required.
Agentic engineering in practice, not just tool usage: you can say where an agent loop earns its keep, what guardrails and structured outputs keep it in bounds, and how you manage its context. Same discipline when coding agents write for you — tests, review gates, repo conventions.
Calm, structured decision-making under tight timelines or ambiguity.
Capable of communicating clearly across engineering, product, and field teams, ensuring alignment from prototype to rollout.
Experienced in spotting risks early and course-correcting without friction when delivery timelines are tight.
Someone who cares deeply about data quality, precision, and governance.
Strong communication and stakeholder-management skills across technical and business teams.

Nice To Haves

Familiarity with metadata systems, data cataloging, or document AI workflows.

Responsibilities

Own end-to-end technical delivery of Unstructured AI systems - from feature prototype to stable production across enterprise environments.
Build and scale full-stack systems that ingest, process, and enrich large volumes of unstructured content from distributed enterprise silos (PDFs, contracts, reports, and other document types).
Shipping complex systems under real ambiguity — often defining the scope and acceptance criteria yourself, not receiving them.
Writing and reviewing production-grade backend code (Python, FastAPI) — you write the services you design, rather than handing implementation to someone else.
Building/deploying document-processing systems that handle large-scale, unstructured data environments.
Integrating data from diverse enterprise data sources (e.g., SharePoint, Salesforce, or internal APIs) to provide context for AI features.
Partnering across engineering, product, and sales teams, ensuring alignment from prototype to rollout.
Occasionally working with modern frontend development.
Drive the development of ambitious, enterprise-grade AI product features that solve for data at scale, architecting the high-performance pipelines and advanced context engineering required to deliver accurate, reliable results.