Senior LLM/RAG Engineer

 (with Graph & Knowledge Graph Expertise)

Key Responsibilities:

LLM & RAG Development:

• Design, implement, and optimize RAG pipelines for domain-specific applications (e.g., chatbots, search engines, enterprise knowledge management).
• Fine-tune and adapt open-source LLMs (Llama 3, Mistral, GPT-neo) for task-specific performance.
• Implement hybrid retrieval systems combining dense (e.g., vector DBs) and sparse retrieval methods.

Graph RAG & Knowledge Graphs:

• Build Graph-enhanced RAG systems leveraging Knowledge Graphs (e.g., Neo4j, Amazon Neptune, RDF) for structured reasoning.
• Develop methods to extract, embed, and query relational knowledge from unstructured/textual data.
• Optimize graph traversals for real-time retrieval and reasoning in RAG workflows.

Engineering & Deployment:

• Write production-grade Python code and APIs (FastAPI, Flask) for LLM serving.
• Optimize pipelines for low-latency inference (quantization, pruning, ONNX runtime).

Collaboration & Innovation:

• Work with cross-functional teams to integrate LLMs into products.
• Stay ahead of SOTA advancements in NLP, graph ML, and multimodal AI.

Required Skills & Qualifications:

• 2+ years of hands-on experience with LLMs, RAG, and Knowledge Graphs.
• Proficiency in Python and NLP libraries (LangChain, LlamaIndex, Hugging Face Transformers).
• Experience with graph databases (Neo4j, TigerGraph, Amazon Neptune) and query languages (Cypher, SPARQL).
• Familiarity with vector databases (Milvus, Weaviate, FAISS) and embedding models (BERT, SBERT).
• Strong understanding of LLM fine-tuning (LoRA, QLoRA, RLHF) and evaluation metrics (RAGAS, BLEU).
• Knowledge of graph ML techniques (Graph Neural Networks, Node2Vec) is a plus.

Preferred Qualifications:

• Master’s/PhD in Computer Science, AI, or related fields.
• Publications or contributions to NLP/Graph ML communities (ACL, NeurIPS, arXiv).
• Experience with multimodal RAG (text + graph + image/video).

Why Join Us?

• Work on groundbreaking AI systems with real-world impact.
• Competitive salary, equity, and flexible work arrangements.
• Collaborative culture with access to cut-edge tools and datasets.