Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
ЗП не указана
На похожих ролях по рынку
$74к$91кмедиана$108к
48 в выборке
На СНГ рынке: $4.2к/мес ($3.2к - $5.1к)
88
Хорошая вакансия
развернуть
Чётко очерченная senior-роль сочетает графы знаний промышленного масштаба с актуальными задачами GraphRAG и AI-агентов, а условия включают широкий пакет льгот; без зарплаты сложнее оценить финансовую сторону предложения.
Кликните для подробной информации
Зарплата не указанаАктуальный стекПонятные задачиAI для устойчивого развитияХороший пакет льготТолько Польша
Оценка от Hirify AI
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Senior Knowledge Graph Engineer (AI): Building production-grade GraphRAG ingestion pipelines, entity-resolution systems, and operational knowledge graphs for autonomous AI agents addressing sustainability challenges with an accent on graph databases, semantic web standards, and hybrid vector-graph retrieval. Focus on optimizing multi-hop queries, integrating enterprise and external data, and implementing deterministic validation and MCP tooling for LLM agents.
Location: Hybrid in Warsaw, Poland, with 2 days per week in the office, or fully remote from Poland. Candidates must be eligible to work and live in Poland.
Company
hirify.global provides business sustainability ratings, combining expert analysis and technology to help companies assess environmental, social, and ethical risks.
What you will do
Design and maintain high-throughput GraphRAG ingestion pipelines for ERP data, SQL sources, ESG reports, and streaming feeds.
Build operational labeled property graphs and RDF triple stores using technologies such as Neo4j, Memgraph, and related platforms.
Develop NER, entity-linking, deduplication, and entity-resolution workflows for vendors, products, facilities, and other enterprise data.
Implement ETL/ELT pipelines that federate supply-chain data with external ontologies, registries, sensor standards, and geospatial data.
Build low-latency GraphRAG retrieval layers, hybrid vector-graph indexes, NL2Query tools, and MCP endpoints for autonomous LLM agents.
Optimize graph queries, partitioning, indexing, and CI/CD validation to support reliable traversal across billions of nodes and edges.
Requirements
4+ years of production experience building and querying labeled property graphs or RDF triple stores.
Advanced Python skills for scalable data pipelines, including experience with RDFLib, NetworkX, or PyGraphistry.
Strong cloud experience, preferably with Azure, Azure Foundry, Azure Bicep, AzureML, and Azure Cloud Storage.
Experience with NLP or LLM-based structured extraction, dbt, vector stores, and hybrid search architectures.
Solid knowledge of RDF, RDFS, OWL, SKOS, SHACL, RDF-star, SPARQL, graph schema design, RML, and R2RML.
Must be eligible to work and live in Poland.
Nice to have
Experience with supply-chain, carbon-accounting, GHG Protocol, or lifecycle-assessment data.
Direct experience building MCP servers for LLM agents.
Experience with enterprise OBDA approaches at scale.
Culture & Benefits
Flexible working hours and hybrid work organization.
Remote work from abroad policy, plus internet and electricity bill allowance.
Office and IT equipment support, wellness allowance, and professional mental health support.
Optional fully covered or co-financed health and life insurance, Multisport card, Multikafeteria, and lunch card.
Learning and development opportunities, sustainability events, community involvement, employee-led resource groups, and peer recognition.
Additional community-service day and referral bonus policy.
Hiring process
Submit a CV in English.
The hiring process can be adjusted to support individual accessibility needs.
Every application receives a response.
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 →