Job Description

1.GenAI Development & Integration
Design and implement GenAI workflows for enterprise use cases.
Develop prompt engineering strategies and feedback loops for LLM optimization.
Capture and normalize LLM interactions into reusable Knowledge Artifacts.
Integrate GenAI systems into enterprise apps (APIs, microservices, workflow engines)
Programming languages: Python
Model Gateway & Multi-LLM Strategy
Architect model gateways to access multiple LLMs (OpenAI, Anthropic, Cohere, etc.).
Dynamically select models based on accuracy vs. cost trade-offs.
Benchmark and evaluate models for enterprise-grade performance.
Agentic Workflows
Design and implement agent-based orchestration for multi-step reasoning and autonomous task execution.
Design and implement agentic workflows using industry-standard frameworks for autonomous task orchestration and multi-step reasoning.
Ensure secure and controlled execution of agentic pipelines across enterprise systems via constraints, policies, and fallback paths.
Data Lakehouse & Knowledge Management
Architect and maintain Lakehouse environments for structured and unstructured data.




Implement pipelines for document parsing, chunking, and vectorization.
Maintain knowledge stores, indexing, metadata governance
Enable semantic search and retrieval using embeddings and vector databases.
Ontology & Taxonomy Engineering
Build and maintain domain-specific ontologies and taxonomies.
Establish taxonomy governance and versioning.
Connect semantic registries with LLM learning cycles.
Enable knowledge distillation from human/LLM feedback.
AI Governance & Knowledge Distillation
Establish frameworks for semantic registry, prompt feedback, and knowledge harvesting.
Ensure compliance, normalization, and promotion of LLM outputs as enterprise knowledge.
Observability & Cost Optimization
Implement observability frameworks for GenAI systems (performance, latency, drift).
Monitor and optimize token usage, inference cost, and model efficiency.
Maintain dashboards for usage analytics & operational metrics.
Make Build vs. Buy decisions based on cost-perk analysis

📌 Staff Data Analytics Engineer Bengaluru
🏢 Sandisk
📍 Bengaluru