Senior Agentic AI Developer

Lester Infoservices; Bengaluru; Hybrid; INR 4,200,000 - 7,000,000; Full time

Job summary

Lester Infoservices in Bengaluru seeks a Senior Agentic AI Developer to design and deploy multi-agent systems and autonomous workflows. The role focuses on generative AI, RAG architectures, and cloud-hosted AI solutions, with emphasis on A2A protocols, memory/planning, and tool use.

You will work with AutoGen, LangGraph, CrewAI, and related frameworks, applying MLOps and orchestration tools to deliver scalable AI products in a hybrid setup.

Qualifications

• 8-12 years of overall technical experience with 4 years in Generative AI/LLM tech.
• 3+ years building agentic systems, workflow automation, or autonomous AI apps.
• Hands-on with AutoGen, LangGraph, CrewAI, or similar agentic frameworks.
• Strong knowledge of workflow orchestration tools (Temporal, Airflow, Prefect).
• Expertise in A2A protocols, MCP, and OpenAPI standards.
• Experience with multi-agent architectures, memory, planning, and tool use.

Responsibilities

• Design, develop, and deploy multi-agent systems and agentic apps.
• Build intelligent workflow orchestration for autonomous decision-making.
• Implement A2A communication and MCP for collaboration.
• Create cloud-hosted Generative AI solutions and RAG workflows.
• Integrate with AWS Bedrock, SageMaker JumpStart, and on-prem infra.
• Tune and evaluate ML models/LLMs for production inference.
• Apply best practices for testing, deployment, and observability.

Skills

Agentic frameworks (AutoGen)

Agentic frameworks (LangGraph)

Agentic frameworks (CrewAI)

Workflow orchestration

A2A protocols

MCP (Model Context Protocol)

RAG architectures

Vector databases

NLP & DL libraries

Python

Cloud platforms (AWS, Bedrock, SageM.)

MLOps/LLMOps

Tools

Docker

Kubernetes

OpenAPI

LangChain

LlamaIndex

Pinecone

Weaviate

Job description

We are looking for a Senior Agentic AI Developer to design cutting-edge AI solutions and autonomous systems focused on agentic workflows and generative AI technologies. This position will be full-time and Hybrid Bengaluru.

What Youll Do

• Design, develop, and deploy multi-agent systems and agentic applications using frameworks like AutoGen, LangGraph, CrewAI, or similar
• Build intelligent workflow orchestration systems that enable autonomous decision-making and task execution
• Implement Agent-to-Agent (A2A) communication protocols and Model Context Protocol (MCP) for seamless agent collaboration
• Develop automation solutions using OpenAPI standards for integration with enterprise systems
• Create self-healing, adaptive workflows that optimize business processes autonomously
• Use ML, deep learning, and Generative AI tools to design, evangelize, and implement state-of-the-art solutions
• Define and implement best practices for building, testing, and deploying scalable AI solutions, with a focus on generative models and LLMs using proprietary or open-source models
• Drive successful business outcomes by designing and building cloud-hosted Generative AI solutions
• Work closely with internal teams to integrate RAG workflows, agent-based systems, and automation frameworks into applications
• Design and implement architectural solutions for Information Retrieval using RAG, Vector DBs, and Knowledge Graphs
• Work with public cloud (AWS) and on-premises infrastructure for deploying LLMs, agents, and orchestration systems
• Evaluate, build, and fine-tune ML models and LLMs to solve complex business problems
• Stay abreast of latest developments in agentic AI, autonomous systems, language models, and generative AI technologies

What You'll Need

Required:

• 8-12 years of overall technical experience with 4 years of hands-on experience in Generative AI and LLM technologies
• 3+ years of experience building agentic systems, workflow automation, or autonomous AI applications
• Deep hands-on experience with agentic frameworks (AutoGen, LangGraph, CrewAI, Agency Swarm, or similar)
• Strong knowledge of workflow orchestration tools and patterns (Temporal, Airflow, Prefect, or similar)
• Expertise in OpenAPI standards, Agent-to-Agent (A2A) protocols, and Model Context Protocol (MCP)
• Experience designing multi-agent architectures with memory, planning, and tool-use capabilities
• Knowledge of agent evaluation, testing frameworks, and observability patterns
• Proven track record of deploying and optimizing LLM models for inference in production environments
• Extensive experience with LLM orchestration frameworks (LangChain, LlamaIndex required)
• Hands-on experience with Amazon Bedrock, SageMaker JumpStart, and other cloud-based LLM platforms
• Expertise in RAG architectures, Fine-tuning techniques, and Prompt Engineering
• Deep understanding of Vector Databases (Pinecone, Weaviate, Milvus, ChromaDB) and Knowledge Graphs
• Expert in NLP techniques and deep learning libraries (Transformer models, LSTM, BiLSTM, CNN, BERT, GPT, T5)
• Proficiency with ML frameworks: TensorFlow, PyTorch, Hugging Face Transformers, scikit-learn
• Strong programming skills in Python (required), plus JavaScript/TypeScript or Node.js
• Deep understanding of data structures, algorithms, and system design patterns
• Hands-on experience in MLOps/LLMOps including data pipelines, model training/refinement, validation, drift management, and serving
• Experience with containerization (Docker, Kubernetes) and CI/CD pipelines for ML systems
• Knowledge of monitoring, logging, and observability tools for production AI systems

Preferred:

• Experience with function calling, tool use, and external API integration in agent systems
• Knowledge of reinforcement learning and agent training methodologies
• Familiarity with semantic reasoning, planning algorithms (ReAct, Chain-of-Thought, Tree-of-Thoughts)
• Experience with graph databases (Neo4j, Neptune) and ontology design
• Contributions to open-source AI/ML projects
• Publications or patents in AI/ML domain