3 дня назад
Middle Strong/Senior AI/ML Engineer (Document Intelligence)
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
Текст:
TL;DR
Middle Strong/Senior AI/ML Engineer (Document Intelligence) (LLM/OCR): Building an AI extraction core for an insurance underwriting platform that classifies mixed-format documents and extracts structured fields with source evidence, with an accent on production AI pipelines, data contracts, and backend integration. Focus on designing provenance-linked extraction, integrating cloud OCR and LLM services through portable adapters, and handling asynchronous jobs, retries, idempotency, and real-world document failure modes.
Location: Work from anywhere
Company
is building a configurable AI product that automates document-heavy submission processing for insurance underwriting teams.
What you will do
- Design canonical data contracts for submissions, source documents, and extracted fields shared by AI and backend services.
- Build OCR and preprocessing pipelines for scanned and native PDF, Office, and other mixed-format documents.
- Implement LLM-based whole-document classification and structured field extraction with exact source evidence and provenance.
- Design cloud-agnostic provider interfaces and adapters for OCR, LLM, and storage services while shipping against one cloud provider first.
- Integrate AI calls into asynchronous queue-based architecture, collaborating with backend engineers on retries, idempotency, and failure handling.
- Implement lightweight logging, tracing, validation, smoke testing, and data structures that support human-in-the-loop underwriting review screens.
Requirements
- 3+ years of experience building applied AI/ML systems in production, ideally for document processing or data extraction.
- Hands-on experience with LLM-based classification and structured extraction, including prompt design, schema-constrained outputs, and quality evaluation on real documents.
- Experience with OCR or document-intelligence services such as Azure Document Intelligence or AWS Textract, including mixed-format document preprocessing.
- Ability to design clean data contracts and schemas for backend and frontend consumers.
- Backend literacy with FastAPI and asynchronous task queues, plus understanding of pragmatic logging, validation, testing, and failure handling.
- Upper-Intermediate or Advanced English required.
Nice to have
- Experience in insurance, underwriting, or another document-heavy regulated domain.
- Experience designing human-in-the-loop review workflows with frontend teams.
Culture & Benefits
- Work-from-anywhere arrangement.
- 18 days of paid annual leave, 10 sick leaves, additional days off, and national holidays.
- Medical care, health check-ups, and a maternity leave bonus.
- Training, conferences, certification support, tech talks, and IT Cluster membership.
- Sports activities, internal events, business trips, and accounting support.
Будьте осторожны: если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого - это мошенники. Обязательно жмите "Пожаловаться" или пишите в поддержку. Подробнее в гайде →