Gen AI Vector DB Engineer

As a Gen AI Vector DB Engineer, you will be responsible for designing, implementing, and optimizing vectorized databases to support the organization's data processing needs. You will work closely with cross-functional teams to ensure efficient data storage, retrieval, and manipulation. The ideal candidate is a seasoned database professional with a strong background in vectorized data processing and a passion for optimizing database performance.

Key Responsibilities:

• Database Designing & Implementation - Design and implement vectorized databases to efficiently store vector embeddings generated by LLMs using vector data bases like Pinecone, OpenSearch or Chroma etc. Ensure seamless integration with applications and analytics platforms.
• Performance Optimization - Assessments of database performance and recommend improvements. Indexing techniques like KD-Trees, Hierarchical Navigable Small World graphs (HNSW), or Inverted Multi-Index (IMI). Fault tolerance using Sharding and Replication.
• Data Modelling - Work on schema design, indexing, and partitioning strategies to enhance data organization using vector embeddings generated by LLM.
• Scalability & Reliability - Architect scalable database solutions that can handle growing data volumes and user loads. Implement strategies for data backup, recovery, and disaster recovery.
• Security & Compliance - Implement and enforce data security measures to protect sensitive information. Ensure compliance with relevant data regulations and industry standards.

Skills & Tools:

• Experience with vector database technologies like Pinecone, Milvus, pgvector and OpenSearch or Elasticsearch etc and experience with graph database (like Neo4j) and knowledge graphs.
• Familiarity with LLM (Large Language Models), both online (OpenAI, Gemini) and offline (Llama2/3, Mistral etc.) concepts related to vector database.
• Knowledge of algorithms for creation of vector index like Random Projection, Product Quantization & Locality-Sensitive Hashing
• Knowledge of various similarity measures (Cosine, Euclidian & dot product) for embeddings & how that impacts performance.
• Strong proficiency in embeddings, vector stores, database optimization, performance tuning, and query languages
• Experience with LLM and related frameworks like Langchain and Haystack
• Programming skills in languages such as Python and any other relevant query languages.
• Excellent problem-solving skills with effective communication & collaboration abilities