About the role

About EXL

EXL is recognized as a global leader in analytics, artificial intelligence, and digital solutions, supporting a wide range of industries through innovative technology and data-driven strategies. Listed on NASDAQ under the ticker EXLS, EXL has a market valuation of approximately $1.3 billion and a global workforce of 48,000+ employees. With a strong presence across continents and a portfolio of 800+ clients worldwide, EXL continues to drive digital transformation at scale through cutting-edge engineering, domain expertise, and customer-centric solutions.

Agentic AI Engineer

Department:	AI Solutions
Location:	Hybrid / Remote
Experience:	5+ Years

About the Role

We are seeking a highly skilled Agentic AI Engineer to design, build, test, and orchestrate autonomous AI agent systems capable of executing complex, multi-step workflows. The ideal candidate will have deep expertise in Large Language Models (LLMs), agent orchestration frameworks, retrieval systems, and enterprise-grade AI deployments.

You will work at the forefront of AI innovation, building production-ready agentic systems that leverage reasoning, planning, tool usage, memory, and human-in-the-loop interactions to solve real-world business challenges.

Core Areas of Expertise

LLM Orchestration
Agent Design & Architecture
Multi-Agent Systems
Tool Integration & Automation
Prompt & Context Engineering
RAG & Knowledge Systems
AI Safety & Governance

Key Responsibilities

Design and develop autonomous AI agents and multi-agent systems with memory, tool usage, fallback strategies, and reasoning capabilities.
Build production-grade agent workflows using frameworks such as LangGraph, AutoGen, CrewAI, Semantic Kernel, or custom orchestration layers.
Implement advanced agent reasoning patterns including ReAct, Plan-and-Execute, Reflection Loops, and Tool-Augmented Generation.
Develop prompt engineering and context optimization strategies to improve reliability, accuracy, and consistency of LLM outputs.
Design and manage orchestration workflows involving task routing, parallel execution, state management, retries, escalation paths, and human-in-the-loop interventions.
Build comprehensive evaluation frameworks including automated testing, adversarial testing, benchmarking, and LLM performance monitoring.
Implement Retrieval-Augmented Generation (RAG) architectures using vector databases, embeddings, semantic search, and knowledge graphs.
Enable observability by tracing agent decisions, tool interactions, execution paths, and LLM calls using monitoring and analytics platforms.
Apply AI security and governance best practices including prompt injection prevention, access control, tool authorization, and safe execution policies.
Optimize agent systems for scalability, latency, throughput, and operational cost.
Develop CI/CD pipelines for AI agents, including version control, testing automation, deployment strategies, and rollback mechanisms.
Integrate AI agents with enterprise applications, databases, APIs, and business workflows.
Design feedback-driven learning mechanisms using production telemetry, evaluation signals, and user feedback.
Leverage Model Context Protocol (MCP) frameworks to enable secure tool orchestration, API integration, and database connectivity.
Fine-tune and adapt foundation models using techniques such as LoRA, QLoRA, PEFT, RLHF, instruction tuning, and supervised fine-tuning.
Stay current with advancements in agentic AI research, frameworks, tooling, and deployment best practices.

Required Skills & Experience

5+ years of overall software/AI engineering experience.
Minimum 3 years of hands-on experience building and deploying LLM-based or agentic AI systems in production environments.
Strong expertise in agentic architecture patterns including ReAct, Reflection, Plan-and-Execute, Tool Calling, and Autonomous Workflow Design.
Advanced proficiency in Python and modern AI development ecosystems.
Hands-on experience with one or more frameworks such as LangChain, LangGraph, AutoGen, CrewAI, Semantic Kernel, or equivalent.
Strong understanding of prompt engineering, context window optimization, structured outputs, and LLM orchestration.
Experience integrating AI agents with external tools, REST APIs, SQL databases, vector databases, code interpreters, and enterprise applications.
Expertise in Retrieval-Augmented Generation (RAG), embeddings, semantic search, vector stores, and knowledge retrieval systems.
Experience designing evaluation frameworks using RAGAS, custom evaluation harnesses, LLM-as-a-Judge methodologies, and benchmarking tools.
Strong understanding of AI safety risks including hallucinations, prompt injection attacks, tool misuse, and mitigation techniques.
Experience with cloud platforms such as AWS, Azure, or GCP.
Hands-on experience with Docker, Kubernetes, and modern DevOps practices.
Familiarity with MLOps and AIOps tools such as MLflow, Weights & Biases, LangSmith, Arize, or equivalent platforms.
Strong experience designing conversational memory systems, semantic caching, context optimization, token management, and cost-efficient AI architectures.

Nice to Have

Experience building large-scale multi-agent systems and inter-agent communication protocols.
Exposure to metadata management, data governance, data lineage, or enterprise data catalog platforms.
Contributions to open-source AI, LLM, or agentic AI projects.
Experience with Java, Scala, PySpark, or COBOL.
Knowledge of graph databases and knowledge graph implementations.
Familiarity with AI observability and governance platforms.

What We Offer

Opportunity to build cutting-edge agentic AI solutions on real-world enterprise data.
Work alongside AI researchers, data engineers, architects, and product leaders.
Exposure to emerging AI technologies and enterprise-scale deployments.
Competitive compensation and benefits.
Flexible work arrangements (Hybrid/Remote).
A collaborative and innovation-driven work culture focused on continuous learning and growth.

Information is subject to change and should be verified from official sources.