Principal / Lead AI ML Engineer – Knowledge Graphs & GenA

Experience Required
10+ years of hands-on experience in AI/ML engineering, with strong depth in knowledge graphs, unstructured data processing, and generative AI systems.

Role Summary
We are seeking a highly experienced AI/ML Engineer with a strong foundation in knowledge graph engineering and generative AI to design, build, and scale intelligent data pipelines that transform large-scale unstructured data into enterprise-grade knowledge graphs.
The ideal candidate will have deep experience in ontology modeling, entity resolution, probabilistic pattern matching, and agentic knowledge base enrichment, combined with strong expertise in LLMs/SMLs, fine-tuning pipelines, and graph-based reasoning systems.
This role involves architecting and delivering production-grade AI systems that integrate LLMs with knowledge graphs, enabling contextual reasoning, anomaly detection, and intelligent automation at scale.
Key Responsibilities
Knowledge Graph & Ontology Engineering
• Design, build, and maintain enterprise-scale Knowledge Graphs from large volumes of unstructured data (text, documents, logs, PDFs, web data).
• Create and evolve ontologies using RDF/OWL, including:
o Entity extraction and linking
o Entity resolution and disambiguation
o Probabilistic pattern matching
o Ontology alignment across heterogeneous data sources
• Implement semantic modeling for complex domains to support reasoning, discovery, and analytics.

Agentic Knowledge Base Enrichment
• Develop agentic AI systems for:
o Automated data gap identification
o Knowledge base enrichment and validation
o Continuous learning and self improving graph pipelines
• Build workflows that combine LLM reasoning with graph traversal and inference.

AI/ML & GenAI Systems
• Design and implement AI/ML pipelines integrating:
o Large Language Models (LLMs)
o Small Language Models (SMLs)
o Reasoning and task specific models
• Build fine tuning pipelines, including:
o Dataset generation and curation
o Training and fine tuning (SFT, PEFT, adapters)
o Evaluation, benchmarking, and deployment
• Apply prompt engineering, RAG, and hybrid LLM + Knowledge Graph (GraphRAG) techniques for contextual intelligence.

Anomaly Detection & Analytics
• Develop anomaly detection systems on top of knowledge graph data at scale.
• Apply graph analytics, embeddings, and ML techniques to detect:
o Semantic inconsistencies
o Behavioral anomalies
o Data quality and relationship drift

Data & ML Engineering
• Build robust data pipelines that ingest, process, enrich, and publish knowledge graph data.
• Implement scalable ML systems using Python for:
o Model development
o Training and tuning
o Inference and deployment

Technical Skills & Expertise
Core AI/ML
• Strong AI/ML engineering background with deep expertise in:
o Python
o Model development, training, tuning, and deployment
• Extensive hands on experience with:
o Large Language Models (LLMs)
o Small Language Models (SMLs)
o Generative AI and reasoning models
o Text generation, summarization, and semantic search workflows
Knowledge Graph Technologies
• Strong experience with:
o Neo4j, GraphDB
o RDF, OWL
o Cypher, SPARQL
• Proven ability to implement:
o Entity linking and resolution
o Semantic search
o Relationship mapping and inference

GenAI Frameworks & Tooling
• Experience building GenAI systems using:
o LangChain, LangGraph
o LlamaIndex
o OpenAI / Azure OpenAI
o Vector databases such as Pinecone and FAISS

MLOps & LLMOps
• Strong experience in MLOps and LLMOps, including:
o MLflow, Azure ML, Datadog
o CI/CD automation for ML systems
o Observability, logging, and tracing
o Model performance monitoring and drift detection
• Experience deploying and operating AI systems in production environments.

Cloud & Scalability
• Experience building and optimizing AI/ML and graph pipelines either of any on:
o Azure
o AWS
o GCP
• Strong understanding of distributed systems, scalability, and performance optimization.
________________________________________
Client is looking for candidates who have experience in building:
• Ontology from large scale data (requires experience in entity resolution, probabilistic pattern matching)
• Agentic knowledge-base enrichment (automated data gap identification, and data enrichment)
• Anomaly detection on top of knowledge graph data at scale
• Fine tuning pipeline (including dataset generation, tuning, evaluation, deployment) for small language models and reasoning models