Knowledge Graph Engineer (AI Development)

About Sentiaflow Sentiaflow is an AI engineering company that places dedicated engineers into client projects — RAG pipelines, LLM integrations, AI agents, and MLOps systems. We need engineers who can be embedded into live client work fast, with minimal ramp-up.

Role Overview We're hiring a Knowledge Graph Engineer to work on client projects requiring GraphRAG, semantic search, entity resolution, and multi-hop reasoning systems. This is a delivery role — you'll be client-facing and expected to produce production-grade work quickly.

What We Want

• Real, hands-on experience with a graph database —

Neo4j preferred , Amazon Neptune / TigerGraph / ArangoDB also acceptable

• Working fluency in Cypher, SPARQL, or Gremlin — not just theoretical knowledge
• Strong Python engineering skills — clean, production-ready code, not notebook scripts
• Direct experience building or integrating GraphRAG pipelines (graph + LLM retrieval)
• Practical understanding of when a knowledge graph outperforms plain vector RAG, and why
• Experience with entity extraction/resolution (NER, relation extraction) on real, messy data
• A portfolio, GitHub repo, or demo that proves this work exists — not just listed on a resume
• Comfort working directly with clients, explaining tradeoffs, and handling ambiguity under deadlines
• Education: B.Tech/M.Tech from IIT or NIT strongly preferred What We Don't Want
• Candidates who list "knowledge graphs" as a keyword but have only used them in a course or tutorial
• Engineers who've only worked with vector databases and are trying to pivot into "GraphRAG" without hands-on graph modeling experience
• No portfolio, no GitHub, no verifiable project — resume claims alone won't be considered

Must-Have Skills (Non-Negotiable)

• Graph database: Neo4j (or equivalent) — hands-on, not just read-about
• Cypher/SPARQL/Gremlin
• Python for data/ML pipelines
• LLM integration experience (LangChain, LlamaIndex, or raw API)
• RAG architecture understanding

Nice-to-Have (Bonus Points)

• Knowledge graph embeddings (TransE, node2vec, GraphSAGE)
• Ontology design (RDF, OWL)
• Hybrid graph + vector DB systems (Pinecone, Weaviate, Qdrant)
• Finance or healthcare domain data experience
• Published work / blog / OSS contribution on GraphRAG

Experience Level

• 3+ years overall software/data engineering experience
• 1+ years specifically on graph-based or knowledge graph systems
• IIT/NIT graduates strongly preferred; exceptional candidates from other institutions with strong verifiable project work will still be considered

Benefits

• Top-of-the-industry compensation — pay benchmarked above market rate for niche AI/graph skill sets
• Performance-linked bonuses tied to client project success
• Exposure to high-value client engagements across finance, healthcare, and enterprise AI
• Fast-track career growth — build a specialized, high-demand skill set (GraphRAG/Knowledge Graphs) that's still rare in the market
• Flexible engagement models (remote/hybrid depending on client)
• Direct mentorship and technical exposure across multiple live production systems, not siloed internal projects
• Opportunity to build a strong portfolio/case-study body of work across diverse client domains