Brillio
Brillio
Brillio ir tehnoloģiju konsultāciju uzņēmums, kas palīdz organizācijām īstenot digitālo transformāciju veselības aprūpes, apdrošināšanas, loģistikas, finanšu pakalpojumu, telekomunikāciju un mediju, dzīvības zinātņu, kā arī mazumtirdzniecības nozarē. Uzņēmuma darbība aptver klientu pieredzi, datus un mākslīgo intelektu, ģeneratīvo mākslīgo intelektu, produktu un platformu izstrādi, kā arī infrastruktūru, mākoņpakalpojumus un drošību. Brillio apvieno inženierzināšanu kompetenci ar nozares zināšanām, lai izstrādātu digitālos produktus un modernizētu tehnoloģiju vidi, palīdzot uzņēmumiem reaģēt uz mainīgajām tirgus un tehnoloģiju prasībām.

Data Scientist, AI Agents and RAG

Build and deploy production AI agents and LLM applications at Brillio, including RAG systems, prompt and retrieval workflows, and evaluation frameworks.

Apraksts

  • Design AI agent workflows from prompt development through production deployment
  • Develop and optimize RAG pipelines, including chunking, embeddings, ranking, and context management
  • Build prompt engineering workflows with systematic testing and chain-of-thought strategies
  • Connect agents to databases, rule engines, validation systems, and formatting tools
  • Establish automated quality checks and validation layers
  • Instrument solutions and work with data scientists on evaluation frameworks
  • Deploy and maintain production AI/ML solutions, with attention to reliability, monitoring, and edge cases
  • Create feedback loops that use expert reviews to improve agent performance

Prasības

  • Bring 2–4 years of experience, including at least 2 years building LLM applications, RAG systems, or AI agent workflows
  • Have advanced proficiency in Python
  • Have hands-on experience with LLM frameworks such as LangChain, LlamaIndex, or similar tools
  • Know prompt engineering and systematic prompt testing
  • Understand RAG architectures, including embedding models, vector stores, retrieval strategies, and re-ranking
  • Have experience building multi-step agent workflows with tool use, branching logic, and robust error handling
  • Have experience deploying and monitoring AI/ML solutions in production
  • Have experience with data pipeline tools and frameworks such as KubeFlow, BentoML, Great Expectations, or Evidently AI

Saistītās vakances

Thomson Reuters

Data Scientist, People Analytics and Machine Learning

Thomson Reuters

Build HR analytics products and machine learning solutions at Thomson Reuters, turning business needs into useful data tools. Develop React and FastAPI applications, Snowflake systems, and LLM-powered workflows for People Analytics.

Atvērt