Description

Job Description

AI Agent Orchestration Engineer

About Us

Majesco isn’t just riding the AI wave – we’re leading it for the P&C, L&AH, and Pension & Retirement markets. Born in the cloud and built with an AI-native vision, we’ve reimagined the insurance and pension core as an intelligent platform that enables insurers and pension & retirement providers to move faster, see farther, and operate smarter. As leaders in intelligent SaaS solutions, we’ve embedded AI and Agentic AI throughout our product portfolio of core, underwriting, loss control, distribution, digital solutions, and pension & retirement administration. Our solutions help customers reimagine their businesses with real-time insights, optimized operations, and enhanced outcomes. At Majesco, you’ll collaborate with experienced leaders and global teams building and scaling AI-native solutions that solve complex, real-world challenges. If you’re looking for a place where your expertise is trusted, your judgment is valued, and your work helps define what’s next for intelligent insurance and retirement technology, Majesco is where that work happens. Job Summary We are looking for an AI Agent Orchestration Engineer to design, build, and scale reliable graph-based AI agents and tools. The role combines strong Python engineering with LangChain and LangGraph expertise, deep understanding of Majesco AI Core architecture, robust tool development, and disciplined prompt engineering. You will work across typed state management, schema validation, graph orchestration, agent and tool registries, API integrations, persistence, and human-in-the-loop workflows. You will help transform complex business requirements into production-ready, observable, and governable agentic experiences. Key Responsibilities - Design and implement typed agent state, schemas, data contracts, and validation using Python and Pydantic. - Build LangChain and LangGraph workflows with conditional branching, parallel task execution, retries, sub-agents, checkpoints, and human-in-the-loop approvals. - Develop and extend Majesco AI Core integrations using GraphExecutor, ExecutorFactory, ToolResolver, Agent Registry, Tool Registry, and MajescoCheckpointer. - Design and implement reusable asynchronous tools, including API integrations, authentication, request/response handling, error recovery, registration, and database metadata. - Create reliable graph-agent execution patterns that support modularity, testability, observability, and operational resilience. - Design, test, version, document, and govern system, routing, tool-use, validation, and response prompts. - Partner with product, engineering, QA, and domain teams to translate business needs into scalable AI-agent capabilities. - Investigate production issues, improve failure handling and execution quality, and contribute to engineering standards for agentic systems. Essential Attributes & Capabilities - Strong Python development skills, including asynchronous programming, type-safe design, testing, and maintainable software practices. - Strong understanding of stateful workflows, schemas, validation, error handling, retries, and distributed task execution. - Ability to reason about complex agent workflows and convert them into clear, modular, and observable graph designs. - Strong analytical and problem-solving skills with a quality-first and “do it right the first time” mindset. - Self-driven, highly motivated, outcome-focused, and comfortable working independently in a remote environment. - Excellent written and verbal communication skills, with the ability to explain technical concepts to technical and non-technical audiences. - Ability to collaborate effectively with multiple teams and work within agile software development cycles. - Strong interest in automation, reusable components, engineering productivity, and continuous improvement. - Sound judgment when designing guardrails, validation logic, human approvals, and fallback behavior for AI systems. Technical Expertise - Python, asynchronous programming, type hints, Pydantic models, schema design, and validation. - LangChain and LangGraph, including state graphs, nodes, edges, conditional routing, parallel execution, retries, sub-agents, checkpoints, and human-in-the-loop workflows. - Majesco AI Core architecture and components, including GraphExecutor, ExecutorFactory, ToolResolver, Agent Registry, Tool Registry, and MajescoCheckpointer. - Asynchronous tool development, API integration, authentication and authorization, structured inputs and outputs, error handling, logging, and resilience patterns. - Tool and agent registration, database metadata, configuration management, and integration with enterprise services. - Prompt engineering across system, routing, tool-use, validation, and response prompts, including prompt testing, versioning, evaluation, and governance. - Unit, integration, and workflow testing for graph-based agent systems. - Git-based development, code reviews, debugging, documentation, and agile delivery practices. Preferred Qualification & Experience - Bachelor’s or master’s degree in Computer Science, Engineering, Information Technology, or a related discipline. - Professional experience building Python-based backend services, workflow systems, AI applications, or agentic solutions. - Hands-on experience developing production-grade applications with LangChain, LangGraph, or comparable orchestration frameworks. - Experience building or integrating LLM-powered applications, including structured outputs, tool calling, prompt evaluation, and guardrails. - Experience with REST APIs, OAuth or token-based authentication, asynchronous service integrations, and relational or document databases. - Experience with observability, automated testing, CI/CD, and troubleshooting distributed or stateful workflows. - Exposure to insurance, financial services, SaaS platforms, or enterprise product development is desirable.