AI / Document Intelligence Engineer

All roles Open role AI / Document Intelligence Engineer Remote Full-time You will build and harden the multi-pass pipeline at the core of Sparengine: page-level vision extraction, document typing, entity and event extraction, and the knowledge graph that serves as the system's truth memory. You care about making a model behave like a cautious technical analyst, not a confident guesser. You are comfortable with Python, LLM orchestration, structured JSON/JSONL data, and fuzzy retrieval. Bonus: prior exposure to OCR pipelines or graph data structures. Full-time, Remote, AI / Document Intelligence Engineer Company Overview Sparengine is an AI-powered aviation technical records reconstruction and audit platform. We process raw maintenance documentation (logbooks, work orders, AD compliance records, CAMP exports) through a multi-pass pipeline that produces structured asset snapshots, component history reports, and pre-purchase inspection deliverables. We operate an international hybrid model with European direction and a high-performance technical team. We prioritize rigor and clarity in a domain where being confidently wrong is the worst possible failure. The Role: Scope & Execution This is the engineering core of the product. You will build and harden the multi-pass pipeline that turns thousands of unstructured pages into a verifiable knowledge graph.

• Build the pipeline: page-level vision extraction, document typing and segmentation, entity and event extraction, and graph construction
• Own the knowledge graph: the system's truth memory, where every claim is attached to evidence with confidence and provenance
• Make the model cautious: enforce the distinction between a task listed on a card and a task actually signed off, so the system behaves like a careful analyst rather than a confident guesser
• Improve retrieval: fuzzy part number and serial number matching, proximity search, and cross-document linking across large dossiers

Technical Requirements

• Python and solid LLM orchestration experience
• Structured data: comfortable with JSON and JSONL at scale, streamable and queryable
• Retrieval: practical understanding of fuzzy matching and context-aware relevance
• A natural instinct for clean code and for catching silent failures before they reach a deliverable

Professional Qualifications

• Proven execution: you have built and shipped real data or AI pipelines beyond tutorials
• Technical maturity: you simplify complex requirements and keep the system reliable
• Bonus: prior exposure to OCR pipelines, vision models, or graph data structures
• Evidence of work: a portfolio or GitHub showing end-to-end systems you have built

Practical Details Location: Remote Type: Full-time Compensation: Based on profile Why Sparengine You will work at the point where messy reality meets structured truth. The product only works if the engineering is honest about what it does not know, so the work rewards people who care about precision, provenance, and shipping systems that experts can trust. You will see your code directly change how high-value aviation assets are bought, sold, and audited. About Sparengine Every aircraft has a hidden story. We make it legible. Sparengine is aircraft intelligence. Decades of scattered records, scans, inspection photos and systems become one connected view of an aircraft: its condition, its history, what the evidence supports and what is missing. Lessors, MROs and traders use it to decide what an aircraft is really worth. A small team in Toulouse and remote. Real problems, real customers, and ownership from day one. Apply for this role Apply for this role

• 1. Your CV and a short conversation
• 2. A reply within 2 business days
• 3. A recorded video, about 14 min
• 4. Meet the team

One application covers every role you are interested in.