GenAI Developer Agentic AI & LLM Platforms

Price Waterhouse Cooper LLP; Chennai; Hybrid; INR 2,600,000 - 4,200,000; Full time

Job summary

Price Waterhouse Cooper LLP is seeking a hands-on GenAI Developer to design, build, test, and deploy Generative AI applications using LLMs, RAG pipelines, and vector search. You will work on enterprise chatbots, knowledge assistants, and document intelligence, integrating with backend systems and databases.

The role requires 5+ years of experience with 2+ years in GenAI/LLM ecosystems, and familiarity with LangChain, LangGraph, CrewAI, and related tools.

Qualifications

• Hands-on GenAI development with LLM APIs and prompt templates.
• Experience building RAG pipelines and vector-based retrieval.
• Knowledge of enterprise document processing and integration with backend systems.
• Familiarity with LangChain, LangGraph, and agentic frameworks.

Responsibilities

• Design, build, test, and deploy GenAI applications end-to-end.
• Develop RAG, semantic search, and knowledge ingestion pipelines.
• Create agentic AI workflows with tool calls and memory components.
• Build backend services and REST APIs for LLM orchestration and data access.
• Perform testing, evaluation, and observability for GenAI apps.

Skills

LLM APIs

Prompt Engineering

Embeddings

Retrieval-Augmented Generation

Semantic Search

Vector Databases

Structured Outputs

Grounding Techniques

LangChain

LangGraph

CrewAI

LlamaIndex

Langflow

Python

FastAPI

Node.js

REST APIs

OpenAPI

Asynchronous Processing

SQL Databases

NoSQL Databases

Vector Databases (tech)

Redis Caching

Object Storage

Azure OpenAI

AWS Bedrock

Google Vertex AI

OpenAI

Hugging Face

Tools

PDF Processing

Word Processing

Excel Processing

HTML Processing

OCR

Document Chunking

Metadata Extraction

Indexing Pipelines

Job description

Job Title: GenAI Developer Agentic AI & LLM Platforms

Location: Chennai

Experience: 5+ years (with 2+ years in GenAI/LLM ecosystems)

Role Overview

We are looking for a hands‑on GenAI Developer to design, build, test, and deploy Generative AI applications using large language models, retrieval‑augmented generation, semantic search, vector databases, prompt engineering, FastAPI‑based services, and agentic frameworks. The role will contribute to enterprise chatbots, knowledge assistants, document intelligence, automation agents, code assistants, and AI‑led business process automation.

Key Responsibilities

GenAI Application Development

• Develop GenAI applications using LLM APIs, prompt templates, structured outputs, RAG pipelines, embeddings, vector search, and tool calling.
• Build conversational AI, enterprise search, document Q&A, summarization, classification, data extraction, and workflow automation use cases.
• Integrate GenAI features with backend applications, portals, APIs, databases, document stores, and enterprise systems.
• Implement prompt orchestration, context management, response formatting, source citation handling, and feedback capture.

RAG, Semantic Search & Knowledge Ingestion

• Build document ingestion pipelines including parsing, OCR coordination, chunking, metadata extraction, embedding generation, indexing, and refresh.
• Implement semantic search and hybrid search using vector databases and search platforms.
• Support retrieval optimization through metadata filtering, chunking strategy, re‑ranking, grounding, and citation generation.
• Work with enterprise knowledge sources such as SharePoint, Confluence, Google Drive, S3, databases, CRM, ITSM tools, and document repositories.

Agentic AI Development

• Develop basic to intermediate agentic workflows using LangChain, LangGraph, CrewAI, LlamaIndex, or equivalent frameworks.
• Implement agents with tools, memory, reasoning steps, API actions, and human‑in‑the‑loop checkpoints.
• Build semi‑autonomous workflows for business process automation while following safety and approval guardrails.

Backend, API & Data Engineering

• Develop backend services using Python, FastAPI, Node.js, or similar technologies.
• Build REST APIs for LLM orchestration, retrieval, ingestion, evaluation, semantic search, and agent execution.
• Work with SQL and NoSQL databases such as PostgreSQL, MySQL, SQL Server, MongoDB, Cosmos DB, DynamoDB, or equivalent.
• Use Redis or equivalent caching for session state, conversational memory, frequently accessed retrieval results, rate limiting, and workflow state.
• Follow secure coding, logging, exception handling, request validation, OpenAPI documentation, and production deployment practices.

Testing, Evaluation & Observability

• Support prompt testing, LLM response validation, RAG evaluation, regression testing, and hallucination checks.
• Use tools such as LangSmith, Langfuse, RAGAS, DeepEval, TruLens, Promptfoo, or equivalent under guidance.
• Capture logs, traces, token usage, latency, cost, feedback, and quality signals for GenAI applications.

Good to Have

• Exposure to Kafka or messaging systems for asynchronous processing.
• Exposure to MCP server development, tool integration patterns, or agent‑to‑tool communication.
• Basic understanding of A2A communication patterns and multi‑agent orchestration.
• Familiarity with Docker, CI/CD, GitHub Actions, Azure DevOps, Jenkins, or equivalent.
• Exposure to fine‑tuning, SLMs, vLLM, Ollama, or Hugging Face model deployment.
• Frontend integration awareness using React, Angular, Streamlit, Gradio, or similar.
• Domain exposure in banking, healthcare, insurance, retail, telecom, or enterprise operations.

Soft Skills

• Strong problem‑solving and analytical thinking.
• Ability to work in agile delivery teams and collaborate with cross‑functional stakeholders.
• Good communication and documentation skills.
• Ability to learn and evaluate fast‑evolving GenAI tools, frameworks, and patterns.

Required Skills

• Core GenAI (Hands‑on)
  
  • LLM APIs
  • Prompt Engineering
  • Embeddings
  • Retrieval‑Augmented Generation (RAG)
  • Semantic Search
  • Vector Databases
  • Structured Outputs
  • Grounding Techniques
• Frameworks (Hands‑on)
  
  • LangChain
  • LangGraph
  • CrewAI
  • LlamaIndex
  • Langflow
• Backend APIs (Hands‑on)
  
  • Python
  • FastAPI
  • Node.js
  • REST APIs
  • OpenAPI
  • Asynchronous Processing
• Data Stores (Working Knowledge)
  
  • SQL Databases
  • NoSQL Databases
  • Vector Databases
  • Redis Caching
  • Object Storage
• Cloud AI (Working Knowledge)
  
  • Azure OpenAI
  • AWS Bedrock
  • Google Vertex AI
  • OpenAI
  • Anthropic Claude
  • Google Gemini
  • Hugging Face
• Document Processing (Working Knowledge)
  
  • PDF Processing
  • Word Document Processing
  • Excel Processing
  • HTML Processing
  • OCR (Optical Character Recognition)
  • Document Chunking
  • Metadata Extraction
  • Indexing Pipelines
• Evaluation & Observability (Basic to Working Knowledge)
  
  • LangSmith
  • Langfuse
  • RAGAS
  • Promptfoo
  • OpenTelemetry Fundamentals
• Security & Responsible AI (Basic Knowledge)
  
  • PII Handling and Data Protection
  • Access Control and Authorization
  • Prompt Injection Prevention
  • Content Filtering
  • Audit Logging
  • Responsible AI Practices and Governance