Software Engineer 3 - GenAI

This position is listed on behalf of a partner company, who manages all applications and next steps. Our partner is looking for a Software Engineer 3 – GenAI based in India. We are looking for an experienced Software Engineer specializing in Generative AI to design, develop, and scale intelligent AI-powered assistants for the energy and utilities sector. In this role, you will take ownership of major components of a production-grade GenAI assistant embedded in web portals and mobile applications. You will build sophisticated multi-agent architectures, develop Retrieval-Augmented Generation (RAG) pipelines, and integrate large language models with enterprise data and APIs. Working closely with GenAI engineers, platform teams, product managers, and QA specialists, you will ensure solutions are scalable, reliable, secure, and cost-efficient. You will also influence technical architecture, establish engineering best practices, and mentor other developers. This is an exciting opportunity to shape next-generation AI experiences and deliver practical innovation in a collaborative, international engineering environment. Accountabilities: • Design, develop, and scale production-ready multi-agent systems, including supervisor and router architectures, specialized agents, LangGraph workflows, tool calling, conversation memory, and checkpointing. • Build robust and strongly typed tool contracts and data-access layers using internal APIs, ensuring that language models interpret and explain verified results rather than generate unsupported numerical information. • Develop and maintain Retrieval-Augmented Generation (RAG) pipelines for utility tariff information and knowledge retrieval, implementing tenant-isolated embeddings and retrieval mechanisms. • Optimize application performance, latency, reliability, and infrastructure costs through caching, prewarming, retries, circuit breakers, and intelligent model routing across different model tiers. • Implement security and AI guardrails to protect applications against prompt injection, unauthorized access, inappropriate scope usage, and tenant data leakage. • Ensure secure data management and privacy across multi-tenant GenAI applications. • Develop observability and evaluation capabilities using tools such as LangSmith to monitor model behavior, agent performance, and application reliability. • Collaborate with Software Development Engineers in Test (SDETs) to create evaluation datasets, regression testing suites, and quality criteria for production releases. • Investigate and troubleshoot production issues using traces, logs, and performance metrics, identifying root causes and implementing long-term solutions. • Support real-time streaming capabilities through WebSocket and Server-Sent Events (SSE), collaborating with frontend teams on seamless widget and application integration. • Lead technical design discussions, architecture reviews, and code reviews while establishing engineering standards and development best practices. • Mentor and support junior engineers, particularly SDE1 and SDE2 developers, helping them improve their technical skills and engineering capabilities. • Communicate technical designs, architectural decisions, implementation trade-offs, and project progress effectively to engineering teams, product stakeholders, and leadership. • Work closely with GenAI engineering, platform, QA/SDET, and product teams to deliver reliable, user-focused AI solutions. Requirements: • 6–10 years of professional software engineering experience, including at least 1–2 years of hands-on experience building and deploying production-grade LLM or Generative AI features. • Strong programming proficiency in Python and good working knowledge of Java. • Practical experience with agent orchestration frameworks such as LangChain, LangGraph, or equivalent technologies. • Hands-on experience integrating Large Language Model APIs, including OpenAI, Anthropic, or Amazon Bedrock. • Strong understanding of prompt engineering, tool/function calling, structured outputs, and LLM application development. • Experience designing and implementing Retrieval-Augmented Generation (RAG) systems using embeddings and vector databases or stores, such as pgvector. • Strong backend engineering and API design capabilities, with experience building scalable, high-performance, and low-latency services. • Practical experience with AWS cloud infrastructure, containers, Kubernetes/EKS, PostgreSQL, Redis, and CI/CD pipelines. • Familiarity with LLM evaluation techniques, LLM-as-a-judge approaches, and AI observability platforms such as LangSmith. • Strong knowledge of system architecture, distributed systems, scalability, and performance optimization. • Understanding of production monitoring, debugging, reliability engineering, and automated testing practices. • Excellent analytical and problem-solving abilities, with the capacity to address complex technical challenges. • Strong communication and collaboration skills, with experience participating in technical discussions and explaining architectural decisions to diverse stakeholders. • Previous experience leading code reviews, contributing to engineering standards, and mentoring junior developers. Nice to Have: • Experience working with time-series data, Internet of Things (IoT), smart meter data, or the energy and utilities industry. • Exposure to voice-based GenAI applications, conversational AI platforms, or Interactive Voice Response (IVR) solutions. • Experience designing and developing multi-tenant Software-as-a-Service (SaaS) applications. • Knowledge of application security, data privacy, tenant isolation, and security enhancements for LLM-powered systems. • Familiarity with ELK Stack and enterprise-level monitoring and logging solutions. Technology Stack: • Programming languages: Python, Java. • GenAI and orchestration: LangChain, LangGraph, OpenAI, Anthropic, Amazon Bedrock. • AI monitori