Overview
Design, build and deploy multi-agent, agent-to-agent (A2A) systems that collaborate, delegate tasks and execute complex business processes across the Elastic IT organisation, using large language models, agent orchestration and A2A integration frameworks.
What you'll do
- Design, build and deploy agent-to-agent (A2A) communication architectures and workflows that enable autonomous agents to collaborate, delegate tasks, and negotiate multi-step business processes.
- Integrate advanced proprietary and open-source LLMs with enterprise APIs using tool-calling methods, and connect with third-party SaaS applications for smooth operations.
- Use advanced Retrieval-Augmented Generation (RAG) techniques, connecting enterprise context through search and vector storage engines so inter-agent operations are accurate and reliable.
- Provision and manage robust cloud environments (AWS, Azure, GCP) using Terraform to support the high-concurrency demands of LLM inference and multi-agent coordination.
- Support modern DevOps practices, including CI/CD pipelines for automated testing, deployment and versioning, and evaluate LLMs and agentic workflows regularly.
- Apply strict security and network fundamentals (VPC configurations, secure API gateways, encryption, IAM controls) to secure inter-agent communication and protect sensitive enterprise data.
- Create systems to monitor how multiple agents interact, check model accuracy, manage token spending, and identify model drift or agent loops.
- Keep detailed technical documentation for LLM integration protocols, A2A interaction flows and cloud infrastructure deployments.
What you'll need
- Experience using AI-assisted development for code review, test generation, working through unfamiliar code, drafting specifications, summarising incidents and accelerating research.
- Experience shipping features across the full stack, from design, build and test to deploy, spanning database and portal, APIs, backend services and the data pipeline.
- Deep experience integrating, fine-tuning and optimising foundation models (e.g., OpenAI, Anthropic, open-source models), including prompt engineering, tool calling and structured outputs.
- Proven success building multi-agent systems, inter-agent messaging pipelines, state management frameworks and dynamic task delegation protocols.
- Hands-on experience with multi-agent orchestration frameworks such as LangGraph, LangChain, AutoGen, etc., along with tracing platforms like LangSmith.
- Strong understanding of emerging AI interoperability standards, such as the Model Context Protocol (MCP) and open multi-agent communication specifications.
- Contribute production code to the product code base using ReactJS, NodeJS, Java/Spring, the Kotlin Multiplatform SDK and its native layers, Angular/TypeScript Portal and JavaScript Web SDK, NoSQL data modelling, and streaming and messaging/queueing technologies.
- Practical knowledge of RAG patterns, vector databases, hybrid search architectures and context management strategies.
- Hands-on experience with DevOps practices and infrastructure automation using Terraform, Docker and Kubernetes for scalable AI deployments.
- Strong knowledge of secure cloud architectures, zero-trust network design, private endpoints and identity management (OAuth, SAML, IAM) for secure A2A interactions.
- Experience setting up observability frameworks (logging, distributed tracing and metrics) to track non-deterministic multi-agent workflows, token costs and LLM output quality.
- Knowledge of the broader agentic and workflow landscape (e.g., Workday A2A, Salesforce Agentforce, ServiceNow AI Agents).
Details
- Location: Bangalore, India.
- Elastic is a distributed company with flexible locations and schedules for many roles.
Read the full description and apply on the company’s own careers page.