GenAI - Application Developer - 11313

LangChain / LangGraph / LlamaIndex

OpenAI / Azure OpenAI / Claude / Gemini / AWS Bedrock

Agent tools, function calling, and workflow orchestration

Model Context Protocol

FastAPI / Flask / Node.js

Docker and basic CI/CD

Cloud basics: Azure / AWS / GCP

LLM evaluation and observability basics

Responsible AI and AI governance awareness

Support low-level design for assigned AI modules.

Design prompt flow, API flow, and response flow for assigned features.

Support RAG design using approved enterprise documents or databases.

Help define agent workflow steps, tools, fallback handling, and human review points.

Keep design simple, secure, and easy to maintain.

Follow architecture guidance and project standards.

Develop GenAI features using Python or other approved technology stack.

Build LLM-based chat, search, summarization, classification, and Q&A features.

Develop RAG pipelines using embeddings, vector search, and retrieval logic.

Create and improve prompts for better response quality.

Build agent workflows that can call tools, APIs, or backend services.

Implement structured outputs like JSON where required.

Write clean, readable, and maintainable code.

Follow coding standards, branch process, and code review comments.

Integrate LLM APIs with application backend.

Connect AI solutions with enterprise systems, APIs, files, databases, and knowledge sources.

Configure vector databases and document retrieval pipelines.

Configure environment variables, model settings, API keys, and service connections securely.

Support tool-use / function-calling implementation for agents.

Support integration with cloud services where needed.

Work with DevOps and platform teams for environment setup.

Test prompts with different user scenarios.

Validate RAG responses against source documents.

Perform unit testing and integration testing for assigned components.

Test agent workflows, tool calls, API calls, and fallback paths.

Validate AI output for accuracy, relevance, safety, and consistency.

Fix defects found during testing and UAT.

Prepare test evidence and validation notes.

Improve prompt quality and reduce unnecessary model calls.

Optimize retrieval logic, chunking, metadata filters, and context usage.

Support response time and token usage optimization.

Tune API calls, retry logic, timeout, and caching where required.

Identify weak responses and suggest improvement actions.

Support cost-aware design and efficient execution.

Follow secure coding and data handling practices.

Use only approved data sources and approved APIs.

Avoid exposing API keys, tokens, passwords, or confidential data.

Support access control and audit logging as per design.

Follow responsible AI guidelines for safe and reliable output.

Add guardrails and validation checks where required.

Escalate data privacy or unsafe-output concerns early.

Support deployment across Dev / Test / UAT / Prod environments.

Prepare code changes for review and release.

Follow Git and CI/CD process as per project setup.

Support release notes and deployment checklist preparation.

Perform post-deployment validation.

Support rollback or quick fix activities when required.

Support production issues related to AI responses, APIs, retrieval, agents, and latency.

Check logs and identify basic failure reasons.

Debug issues related to wrong answers, missing context, tool failure, or API errors.

Provide RCA inputs for recurring issues.

Implement fixes with proper testing.

Support hypercare after production release.

Prepare technical notes for assigned AI components.

Document prompt behavior, API usage, RAG flow, tool flow, and configuration steps.

Maintain test cases and validation results.

Update support notes and runbooks where required.

Share implementation details with team members.

Support knowledge transfer to QA, support, and delivery teams.

Work in Agile/Scrum delivery model.

Participate in daily stand-ups, sprint planning, reviews, and retrospectives.

Provide clear daily updates on progress, blockers, and next steps.

Work closely with AI architects, senior developers, QA, business analysts, and DevOps teams.

Take ownership of assigned stories and deliver on time.

Raise risks and blockers early.

Cloud / Platform: Azure / AWS / GCP

AI Tools: OpenAI, Azure OpenAI, Claude, Gemini, AWS Bedrock, local/open-source LLMs

Agentic AI Tools: LangChain, LangGraph, LlamaIndex, Agents SDK, MCP

Data Tools: Vector databases, embeddings, document parsers, RAG pipeline

Database / Warehouse: PostgreSQL, SQL Server, MongoDB, Vector DB

Programming Languages: Python, JavaScript / TypeScript, SQL

API / Backend: FastAPI, Flask, Node.js, REST APIs

DevOps Tools: Git, GitHub, Azure DevOps, Docker, CI/CD basics

Monitoring Tools: Application logs, API logs, cloud monitoring, LLM evaluation logs

Documentation Tools: Jira, Confluence, SharePoint, Azure Boards