Job Objective
Support the development of AI products using LLMs with Retrieval-Augmented Generation (RAG), including building retrieval pipelines, evaluating quality, and optimizing for practical use.
Knowledge
- Understands embedding models (differences in dimensions, domains, multilingual capabilities).
- Knows how to evaluate RAG quality (recall@k, precision, faithfulness, hallucination rates).
- Understands prompt engineering strategies for better integration between retrieval and generation
Skills
- Implements custom retrieval strategies (e.g., hybrid search = dense + sparse).
- Tunes chunking, embedding, retrieval hyperparameters (window size, top_k).
- Can handle structured + unstructured data sources (SQL, APIs, documents).
- Integrates reranking models (like Cohere Rerank, bge-reranker).
Example tasks
- Build a chatbot that combines FAQs, SQL queries, and PDF knowledge.
- Experiment with different embeddings (e.g., OpenAI vs InstructorXL vs bge).
- Add caching & latency optimization for production use.
What are the benefits we offer?
● Working hours: from 8.30 am – 5.30 pm, Monday – Friday
Job Type: Full-time
Pay: 7,000,000₫ - 10,000,000₫ per month
Report job
