GenAI Engineer - Agentic ERP Platform in India at Jobgether

This position is listed on behalf of a partner company, who manages all applications and next steps. Our partner is looking for a GenAI Engineer – Agentic ERP Platform based in India.

This role sits at the cutting edge of enterprise AI, focused on building intelligent agent systems that interact directly with large-scale ERP platforms such as SAP, Oracle, and JD Edwards. You will design and develop AI agents that automate complex enterprise workflows, reason over structured and unstructured business data, and support real-time decision-making. The position blends advanced LLM engineering, prompt and context design, and production-grade software development. You will work on defining how AI agents think, behave, and execute tasks safely within enterprise environments. This is a highly experimental yet production-critical role, where innovation must be balanced with reliability, security, and enterprise-grade performance. You will collaborate with engineering, product, and platform teams to build scalable AI-driven ERP capabilities. The environment is fast-evolving, deeply technical, and centered on redefining how enterprises interact with ERP systems through agentic AI.

• Design and build AI agents using Python and frameworks such as Pydantic AI to automate ERP workflows and enterprise business processes.
• Implement MCP-based tool integrations enabling agents to interact with ERP systems, databases, and external enterprise services.
• Develop multi-step agent workflows with decision branching, error handling, and human-in-the-loop escalation logic.
• Engineer system prompts, templates, and guardrails that define agent behavior, constraints, and enterprise-safe responses.
• Build and optimize context engineering strategies, including memory systems, retrieval-augmented context, and token-efficient summarization.
• Integrate and orchestrate LLMs using gateways such as LiteLLM, ensuring routing, fallback handling, and performance optimization.
• Develop evaluation frameworks to measure agent accuracy, reliability, and task success rates across enterprise use cases.
• Implement safety, security, and compliance guardrails, including prompt injection detection, audit logging, and output validation.

• Bachelor’s or Master’s degree in Computer Science, Artificial Intelligence, or a related field.
• 5–8 years of software engineering experience, including at least 2+ years in GenAI, LLM, or AI-driven application development.
• Strong Python expertise with production-level experience in async programming and scalable system design.
• Hands-on experience building LLM-powered applications, agents, or chatbots in production environments.
• Deep understanding of prompt engineering, including system prompts, few-shot learning, structured outputs, and evaluation methods.
• Experience with LLM APIs (OpenAI, Anthropic, or similar), including tool calling, streaming, and structured responses.
• Familiarity with agent frameworks such as Pydantic AI, LangChain, or LlamaIndex.
• Knowledge of context window optimization, token management, and LLM limitations.
• Experience with API integrations, REST/JSON systems, and enterprise software architectures.
• Strong debugging skills for LLM behavior, failure modes, and performance tuning.
• Preferred: experience with MCP, LiteLLM, RAG systems, vector databases, workflow engines (Temporal/Restate), and ERP systems (SAP/Oracle).

• Opportunity to build next-generation agentic AI systems for enterprise ERP transformation.
• Highly technical, innovation-driven environment focused on cutting-edge GenAI applications.
• Exposure to large-scale enterprise systems and global ERP platforms.
• Strong focus on research-driven engineering, experimentation, and AI system design.
• Remote-friendly setup with collaboration across global engineering teams.
• Career growth in advanced AI engineering, agent frameworks, and LLM systems architecture.
• Opportunity to work on impactful AI products used in mission-critical enterprise workflows.