GEN AI_Noida_Hybrid

Key Responsibilities

• Develop GenAI applications using Python, LLM APIs, prompt engineering, RAG patterns, embeddings, vector search, and agentic AI frameworks.
• Build AI agents capable of reasoning, planning, tool calling, function calling, workflow orchestration, memory usage, task decomposition, and multi-step execution.
• Design and implement RAG and Agentic RAG pipelines using document ingestion, parsing, chunking, metadata tagging, embeddings, vector indexing, semantic search, hybrid retrieval, reranking, prompt construction, and grounded response generation.
• Integrate GenAI solutions with structured and unstructured enterprise data sources such as documents, databases, SharePoint repositories, knowledge bases, APIs, ticketing systems, and workflow platforms.
• Implement prompt templates, system prompts, reusable prompt libraries, structured outputs, JSON response formats, prompt versioning, and output validation logic.
• Create tool integrations that allow agents to call APIs, execute workflows, retrieve data, summarize content, classify information, generate reports, and trigger downstream actions safely.
• Support model selection and configuration based on use case needs such as accuracy, latency, context window, token usage, cost, privacy, and deployment constraints.
• Integrate LLMs with enterprise systems, APIs, databases, knowledge repositories, search services, and automation workflows.
• Use frameworks such as LangChain, LangGraph, LlamaIndex, Semantic Kernel, AutoGen, CrewAI, or similar tools to build agentic workflows.
• Create reusable components for prompt templates, tool integrations, retrieval workflows, memory handling, guardrails, model evaluation, tracing, observability, and monitoring.
• Perform testing and evaluation of GenAI outputs for factual accuracy, relevance, hallucination control, groundedness, safety, bias, latency, token usage, cost, and reliability.
• Implement basic Responsible AI and security controls such as input validation, prompt injection mitigation, PII handling, access control, auditability, and safe response handling.
• Support deployment and integration of GenAI components into backend services, web applications, automation workflows, and cloud-hosted environments.
• Support testing and evaluation of GenAI outputs for accuracy, relevance, hallucination control, latency, cost, safety, and reliability.
• Collaborate with data engineers, backend developers, cloud engineers, QA teams, product owners, and business stakeholders to deliver AI-enabled features.
• Document prompts, agent flows, API integrations, model configurations, assumptions, limitations, and deployment steps.

Skill RequirementsMandatory Technical Skills

• Strong programming capability in Python, including data structures, APIs, object-oriented programming, exception handling, logging, debugging, package management, and modular application development.
• Hands-on exposure to Generative AI, Large Language Models, prompt engineering, embeddings, tokenization, context windows, structured outputs, and AI application development.
• Working knowledge of RAG architecture, including document processing, chunking strategies, metadata design, vectorization, semantic search, hybrid search, reranking, context augmentation, and grounded response generation.
• Experience or strong project exposure in agentic AI concepts such as tool calling, function calling, planning, memory, reflection, reasoning loops, task decomposition, autonomous execution, human-in-the-loop flows, and workflow orchestration.
• Exposure to frameworks such as LangChain, LangGraph, LlamaIndex, Semantic Kernel, AutoGen, CrewAI, or equivalent agentic development frameworks.
• Experience integrating LLMs through APIs or cloud AI services such as AWS Bedrock, Azure OpenAI, Google Vertex AI, OpenAI APIs, Anthropic APIs, or open-source model endpoints.
• Working knowledge of vector databases or semantic search platforms such as FAISS, Pinecone, Chroma, Weaviate, OpenSearch, Azure AI Search, or pgvector.
• Understanding of REST APIs, JSON, authentication, backend integration, data ingestion, and application deployment basics.
• Working knowledge of Git, code reviews, debugging practices, documentation, and Agile delivery methods.

Preferred / Additional Skills

• Exposure to cloud-native AI services, especially AWS Bedrock, Amazon SageMaker, Azure OpenAI, Azure AI Search, Google Vertex AI, or Gemini APIs.
• Familiarity with multi-agent systems, supervisor-agent patterns, planner-executor workflows, human-in-the-loop flows, and agent evaluation methods.
• Knowledge of LLMOps or GenAIOps practices, including prompt versioning, model configuration management, monitoring, tracing, evaluation, and cost tracking.
• Exposure to OCR, document intelligence, NLP, text classification, summarization, entity extraction, and enterprise knowledge search use cases.
• Basic understanding of model evaluation metrics, hallucination reduction, grounded responses, guardrails, responsible AI, privacy, and security controls.
• Experience with FastAPI, Flask, Streamlit, React, Node.js, or similar tools for building AI-enabled applications and demos.
  
• Must Have Skills
• Click Enter to show the proficiency description of Core RPA - UiPath/BP/AA/PACore RPA - UiPath/BP/AA/PA
• Click Enter to show the proficiency description of Verbal CommunicationVerbal Communication
• Click Enter to show the proficiency description of AlgorithmsAlgorithms
• Click Enter to show the proficiency description of Learning AgilityLearning Agility
• Click Enter to show the proficiency description of Written Written Communication

Good to have Skills

• Click Enter to show the proficiency description of GenAIExperience Criteria1 to 4 years of relevant experience in GenAI development, AI application engineering, Python development, ML/NLP application development, backend development, or automation engineering.

Educational QualificationsMandatory Qualification:

• B.E. / B.Tech in Computer Science, Information Technology, Artificial Intelligence, Data Science, Electronics, Software Engineering, or any other relevant engineering stream.