Data Scientist – GenAI / LLM / Conversational AI
Role
Data Scientist – GenAI / LLM / Conversational AI
Experience
6+ years total in AI/ML, Data Science, or Software Engineering, including 3+
years focused on GenAI, LLM, RAG, or Conversational AI
Cloud Requirement
Minimum 4 years of hands-on Google Cloud Platform (GCP) experience
Locations
Pune | Hyderabad | Bangalore
Employment Type
Full-time
About the Role
We are looking for an experienced Data Scientist to design, build, and
productionize Generative AI and LLM-powered applications on Google Cloud
Platform. The ideal candidate combines strong software engineering
fundamentals with hands-on expertise in LLM application development,
Retrieval-Augmented Generation (RAG), agent orchestration, and
enterprise-grade LLM security, delivering scalable and secure conversational
AI solutions.
Experience & Qualifications
-
6+ years of overall experience in AI/ML, Data Science, or Software
Engineering.
-
3+ years of hands-on experience in GenAI, LLM, RAG implementation, or
Conversational AI, including production deployment (LLM
productionisation).
-
Minimum 4 years of hands-on experience on Google Cloud Platform (GCP).
Must-Have Skills — Generic
-
Strong programming skills in Python, with experience building APIs using
FastAPI and REST API design.
-
Experience with asynchronous processing and building scalable backend
services.
-
Hands-on LLM application development experience.
-
RAG (Retrieval-Augmented Generation) implementation using vector
databases.
-
Strong grounding in prompt engineering, tool calling / function calling,
and structured output generation.
-
Working knowledge of LLM security practices, including prompt injection
prevention, data leakage protection, access control, and guardrail
implementation.
Must-Have Skills — GCP
-
Vertex AI & Vector Search:
Hands-on experience with Vertex AI, including Vector Search, or equivalent
vector database platforms such as Pinecone, Weaviate, FAISS, Chroma, or
pgvector.
-
Gemini:
Practical experience building applications using Google's Gemini models.
-
Agent Orchestration:
Experience orchestrating agents using ADK (Agent Development Kit), or
equivalent frameworks such as LangChain, LlamaIndex, Semantic Kernel,
CrewAI, or AutoGen.
-
Deployment:
Proven experience deploying production workloads on Cloud Run, GKE, or
equivalent container/serverless platforms.
-
Evaluation:
Experience with LLM evaluation using Vertex AI Evals, RAGAS, custom
evaluation frameworks, golden datasets, or regression testing
methodologies.
Good to Have (Trainable)
-
Google Agent Development Kit (hands-on depth).
-
Agent Engine / Gemini Enterprise Agent Platform.
-
Model Armor for LLM security and safety enforcement.
-
Agent observability and tracing tooling.
-
Multi-agent architecture design and implementation.
-
Human-in-the-loop (HITL) approval workflows.
-
Enterprise knowledge graph or enterprise search integration.