Sr AI Engineer
- AIML
- GEN AI
- RAG
- LLM
- NLP
- CI/CD
- Graph Store
- ontology
- Design, build and maintain domain ontologies and taxonomies (classes, properties, relationships, constraints) in OWL/RDF or property-graph models.
- Work with business and domain experts to capture concepts, resolve ambiguity and keep a shared vocabulary.
- Build and evolve knowledge graphs, including entity extraction, entity resolution and linking from structured and unstructured sources.
- Define versioning, governance and validation rules (for example, SHACL) so the ontology stays consistent as it grows.
- Design and ship end-to-end RAG pipelines: ingestion, chunking, embeddings, indexing, retrieval, re-ranking and generation.
- Combine vector search with knowledge-graph retrieval (GraphRAG / hybrid retrieval) to improve answer accuracy and traceability.
- Use the ontology to drive metadata enrichment, query expansion and filtering in retrieval.
- Engineer prompts and context assembly, and add guardrails against hallucination and data leakage.
- Set up offline and online evaluation for retrieval and generation (for example, recall@k, faithfulness, answer relevance) and act on the results.
- Deploy, monitor and optimize services for latency, cost and reliability.
- Write clean, tested, secure code and review the work of peers.
- Mentor junior engineers and share ontology and RAG best practices across teams.
- 3–4 years of professional experience in AI/ML engineering, NLP or data engineering, with at least one production LLM or RAG system shipped.
- Bachelor's or Master's degree in Computer Science, AI, Data Science or a related field, or equivalent experience.
- Hands-on ontology work: modeling with OWL, RDF/RDFS and SKOS, querying with SPARQL, and tools such as Protégé.
- Experience with at least one graph store: Neo4j (Cypher), GraphDB, Stardog, Amazon Neptune or similar.
- Strong RAG knowledge: chunking strategies, embedding models, vector databases (for example, pgvector, Pinecone, Weaviate, Milvus, FAISS), hybrid search and re-ranking.
- Practical experience with LLM APIs and frameworks such as LangChain, LlamaIndex or equivalent.
- Strong Python skills, plus REST API development (for example, FastAPI) and SQL.
- Familiarity with evaluating RAG systems and debugging retrieval quality.
- Working knowledge of Git, Docker and CI/CD.
- Experience with GraphRAG, knowledge-graph-augmented generation or agentic retrieval workflows.
- Exposure to schema.org, industry-standard vocabularies, or retail and e-commerce product taxonomies.
- Knowledge of information-extraction techniques (NER, relation extraction) using LLMs or classical NLP.
- Experience with RAG evaluation tools such as RAGAS, TruLens or DeepEval.
- Cloud experience (GCP, AWS or Azure), including managed AI and data services.
- Understanding of data privacy, access control and secure handling of sensitive data in AI systems.