Revenue Tech Engineering - Software Engineer II - AI

Overview

Elastic, the Search AI Company, enables everyone to find the answers they need in real time, using all their data, at scale — unleashing the potential of businesses and people. The Elastic Search AI Platform, used by more than 50% of the Fortune 500, brings together the precision of search and the intelligence of AI to enable everyone to accelerate the results that matter.

By taking advantage of all structured and unstructured data — securing and protecting private information more effectively — Elastic’s complete, cloud-based solutions for search, security, and observability help organizations deliver on the promise of AI.

We are looking for a Software Engineer to join our Foundational AI team. Our mission is to build the core AI experiences that power both our internal Field teams and our customer-facing support agents and chatbots. Built on Elasticsearch's vector search capabilities and Elastic Observability, our AI systems have a foundation most teams can only dream of—and you'll help us take full advantage of it.

We believe that we can fundamentally transform how our users interact with our platform. You will be a key part of this journey, moving beyond simple prompt engineering to build robust, agentic systems that interface with real-world data and APIs. We are looking for engineers who are excited by the challenge of making AI reliable, observable, and deeply integrated into our business workflows.

This posting reflects our current stack, not a forever decision. AI tooling evolves fast, and we're willing to evolve with it—when the juice is worth the squeeze. Bring your best thoughtful ideas, and let's build better together.

• Architecting Agentic Workflows: Design and implement agentic workflows using Lang Chain and Lang Graph, "Tool Use" (function calling) logic that allows LLMs to interact reliably with internal data sources and third-party APIs.
• Prompt Engineering & Orchestration: Craft, test, and version-control complex system prompts to ensure consistent, high-quality responses across multiple AI personas.
• Building Ingestion Pipelines: Own the data lifecycle for our AI features by building and maintaining scalable ingestion workflows using Apache Airflow.
• Advancing RAG Architectures on Elasticsearch: Leverage Elasticsearch's vector search and hybrid retrieval capabilities to optimize how we chunk, embed, and retrieve information to minimize hallucinations and maximize relevance.
• Ensuring AI Reliability: Implement observability and tracing Elastic APM and Lang Smith to monitor agent performance, identify failure points in tool calls, and ensure security guardrails.

You don't need to check every box—if you're strong in AI/LLM fundamentals and eager to learn the rest, we want to talk.

• AI & LLM Proficiency: Hands-on experience working with LLM APIs and a deep understanding of prompt engineering and model behavior. We currently use OpenAI, Anthropic, Lang Chain, and Lang Graph. Experience with other providers (Google Vertex AI, AWS Bedrock, Cohere), open-source models (Llama, Mistral), or other orchestration frameworks (Llama Index, Semantic Kernel, Haystack) works too.
• Backend Expertise: Strong proficiency in Python for AI logic and pipeline development, with a proven track record of writing clean, maintainable code.
• Language Versatility: Professional experience with an object-oriented language beyond Python (Type Script, Java, C#, etc.) in Type Script, with the ability to build and interface with modern web services beyond just script-based automation. We use Type Script, but this is not a primary focus of the role.
• Data Engineering Fundamentals: Experience building and scheduling data pipelines  currently use Apache Airflow and Elasticsearch. Experience with other orchestration tools (Dagster, Prefect, Temporal), streaming systems (Kafka, Flink), or data warehouses (Snowflake, Big Query, Databricks) works too.
• Search & Retrieval: Familiarity with search fundamentals and vector-based retrieval. We use Elasticsearch for lexical and vector search, along with many other features of the Elastic Stack.