20 hours ago
Bengaluru, IndiaSenior
Responsibilities
- Design and build agentic systems that plan, reason, call tools, and orchestrate multi-step workflows.
- Develop and deploy agents using agent orchestration frameworks such as Google ADK, LangChain/LangGraph, Semantic Kernel, CrewAI, and AutoGen.
- Implement tool calling, function calling, and Model Context Protocol integrations with internal APIs, databases, and third-party services.
- Architect memory, state management, and context handling for long-running and multi-turn interactions.
- Build RAG pipelines and integrate vector databases where appropriate.
- Evaluate, fine-tune, and select foundation models based on cost, latency, and accuracy.
- Implement guardrails, evaluation frameworks, tracing, logging, and observability for agent reliability and safety.
- Collaborate with product, design, and domain experts to translate business workflows into agent capabilities.
- Write production-grade Python or TypeScript code and follow testing, deployment, and documentation practices.
Requirements
- At least 5 years of experience is required.
- At least 3 years of software engineering experience, including at least 1 year working directly with LLMs or generative AI systems in production.
- Hands-on experience with an agent framework such as Google ADK, LangChain/LangGraph, CrewAI, AutoGen, Semantic Kernel, or a similar toolkit.
- Strong production-quality Python skills; async programming experience is a plus.
- Practical knowledge of prompt engineering, tool and function calling, and multi-agent orchestration.
- Experience integrating OpenAI, Anthropic, Google Gemini, or similar LLM APIs and managing rate limits, token economics, and context constraints.
- Familiarity with vector databases and RAG architecture.
- Understanding of evaluation methods for non-deterministic AI systems, including offline evaluations, human review, and A/B testing.
- Experience deploying services on GCP, AWS, or Azure using Docker and Kubernetes.
- Understanding of API design, microservices, and event-driven architectures.
- Preferred experience with MCP server or client implementations, Vertex AI Agent Builder, workflow orchestration tools, and LLM observability or tracing tools.
Tech Stack
Categories
About Accenture
Accenture is a global professional services firm providing management consulting, systems integration and technology, cybersecurity, and business process outsourcing for enterprises and governments. It operates a services-driven model delivering projects and managed services, often with major cloud and software partners, across industries worldwide. Headquartered in Dublin and publicly traded on the NYSE (ACN), it originated as Andersen Consulting and adopted the Accenture name in 2001.