5 жовт. · Sigma Software · тип компанії не вказано
Понад 7 днів вакансія не з’являлася у видачі джерела — можливо, її вже закрито
Завантажте CV — і ми покажемо, що збігається, чого бракує і чи проходить вакансія ваші умови.
Відкрити на DjinniYou will join one of two AI initiatives: ChatCPI, an LLM-powered assistant for intelligent search and question answering over large technical documentation, or a Knowledge Graph Platform focused on structuring and connecting enterprise information for AI-driven use cases.
Customer Our Customer is a leading Swedish telecommunications company recognized for delivering innovative large-scale technology solutions across complex infrastructure environments. The company operates internationally, serving millions of users and working with global technology leaders to drive digital transformation in the telecommunications industry.
Project You will join one of two AI initiatives: ChatCPI, an LLM-powered assistant for intelligent search and question answering over large technical documentation, or a Knowledge Graph Platform focused on structuring and connecting enterprise information for AI-driven use cases. Technologies: Python, LLMs, RAG, vector embeddings, hybrid search, knowledge graphs.
Requirements 3+ years of commercial software development experience 2+ years of experience working on AI/ML projects Good Python skills with hands-on implementation experience Practical experience with LLMs and modern AI technologies Practical experience with RAG architecture, vector embeddings, and hybrid search Experience with information processing aimed at improving AI quality, including data preparation, chunking, metadata enrichment, and relevance tuning Ability to work independently and take ownership of tasks from design through deployment Upper-Intermediate English
Experience with knowledge graphs or graph databases Background in the telecommunications domain
Responsibilities Design and implement LLM-based features and RAG pipelines from data ingestion to answer generation Build and optimize retrieval using vector embeddings and hybrid semantic and keyword search Process, clean, and structure source information to improve the accuracy and relevance of AI-generated output Evaluate model quality, experiment with new approaches, and bring proven solutions into production Work closely with the team to integrate AI components into the wider platform Own changes end-to-end, including design, implementation, testing, and deployment Work independently while proactively communicating blockers, scope questions, and technical trade-offs