Senior GenerativeAI Engineer

Role Summary We are seeking a seasoned, production-focused

Senior Generative AI Engineer to architect, build, and maintain the entire operational lifecycle of our enterprise-grade AI applications. In this position, your value is measured not by simple baseline prompt designs or standalone local prototypes, but by your ability to bridge raw data systems, orchestration logic, specialized evaluation layers, and live cloud deployment engines into an integrated, self-optimizing system. The ideal candidate treats LLM application engineering as an iterative, continuous software lifecycle. You will have direct engineering ownership over real-world data ingestion pipelines, highly optimized advanced retrieval networks, complex multi-step reasoning capabilities, and post-production execution telemetry.

Skills & Qualifications (Must Have) Candidates must satisfy all of the following requirements to be considered:

Experience Foundations:

• 5 to 8 years of overall professional, hands-on software development experience.
• Minimum

2 years of deep, dedicated engineering experience working directly with foundation models, vector spaces, and semantic routing architectures.

• Proven track record of taking at least

one enterprise GenAI or Retrieval-Augmented Generation (RAG) platform completely into production with active user traffic.

Ingestion & Data Phase:

• Advanced mastery of Python development, complex SQL data structures, and multi-tenant database designs.
• Demonstrated ability parsing, handling, and sanitizing large-scale heterogeneous document corpuses.
• Mandatory implementation experience using advanced document parsing and chunking methodologies including

Parent-Child Retrieval, Semantic Context Chunking, and Hierarchical Structuring .

Context, Model Orchestration & Advanced Retrieval:

• Deep theoretical and practical knowledge of context window limits, token boundaries, sub-word tokenization algorithms, context-window saturation, and cost/latency balance boundaries.
• Production orchestration experience using dominant framework stacks:

LangChain, LangGraph, or Semantic Kernel .

• Production-Grade Advanced RAG: Verifiable experience implementing advanced, robust semantic layers including Hybrid Search (Lexical + Dense Vector), Vector metadata filtering, and semantic re-ranking systems (e.g., Cohere Re-rank, BGE Cross-Encoders).
• Deep integration competency managing cloud vector engine stores (e.g., Pinecone, ChromaDB, Weaviate, Qdrant, or Azure AI Search).
• Functional engineering with API-driven tool interfaces, runtime function-calling patterns, and multi-agent cyclic execution paths.

System Evaluation, Guardrails & Governance:

• Mandatory architectural experience constructing automated offline or online evaluation matrices using dedicated frameworks such as

RAGAS, DeepEval, Promptfoo, or LangSmith .

• Proven understanding of metrics quantifying system performance: Faithfulness, Answer Relevance, and Context Precision.
• Hands-on programming of preventative structural guardrails handling prompt injections, data jailbreaks, and real-time automated PII masking layers.

Scale Cloud Deployment & LLMOps:

• Experience deploying enterprise microservices through secure endpoint layers via Azure OpenAI, Amazon Bedrock, or native OpenAI enterprise gateways.
• Mandatory containerization workflow expertise (

Docker ) integrated with enterprise cloud orchestration targets (

Kubernetes, AKS, Azure Container Apps, or AWS ECS ).

• Proven implementation of persistent production runtime observability, logging token usage metrics, rate-limiting handlers, latency spike isolation, and system telemetry loops.

Preferred & Nice-To-Have Skills:

• Complex Orchestration: Deep knowledge of asynchronous execution loops and state machines for autonomous processing utilizing LangGraph, CrewAI, or AutoGen.
• Big Data Systems: Exposure to PySpark, automated distributed ingestion frameworks, and real-time vector indexing pipelines.
• Enterprise Identity & Security: Setting up Role-Based Access Control (RBAC) dynamically mapped straight down into vector record metadata filters.
• Certification: Certification and/or coursework in AI/ML and Generative AI, including platforms such as TensorFlow and AWS Certified Machine Learning – Specialty.

Key Responsibilities Across the GenAI Lifecycle:

• Architect & Code: Develop clean, asynchronous, highly performant microservice backends in Python to expose semantic APIs to scaling consumer layers.
• Optimize Semantic Relevance: Continuous manipulation of vector spaces, metadata indexes, and re-ranking algorithms to drive precision up and decrease hallucinatory rates.
• Establish Safety Frameworks: Architect defensive execution blocks on all incoming and outgoing model tokens to actively mitigate vulnerabilities and handle data exposure boundaries.
• Drive Lifecycle Telemetry: Instrument deep analytics layers to record real-world user interactions (e.g., thumbs-up/down qualitative flags). Process incoming telemetry metrics back to drive systematic adjustments in chunking layouts or foundational prompting styles.

Work Location: Whitefield, Bangalore