Senior AI Engineer, Agentic Systems

Rubrik secures and accelerates the world’s AI transformation. As AI agents reshape the enterprise, Rubrik’s Enterprise AI Incubation team proves it on Rubrik itself.
We are building a thinking enterprise: applications that don't just record the business but help run it. They learn from our data, act through our own agents, and improve with every decision.
Fully backed by executive leadership, this team drives meaningful, large-scale enterprise transformation.

We want senior engineers who have already shipped agents that real people rely on, and who are ready to do it again at higher stakes. You will drop into the real work, redesign it around agents that carry it end to end, and ship it as the new way that part of Rubrik operates.You’ll own your agent systems end to end: understanding the workflow, designing the architecture, writing the code, deploying, and running it in production. Rather than simply querying an API, you engineer robust, production-grade infrastructure around model endpoints, integrating custom tooling, context pipelines, automated evaluations, and defensive guardrails to ensure resilience against unstructured real-world data.

You also decide whether a problem needs an agent, a simpler automation, or nothing at all, and you have explicit authority to stop or re-scope work that won’t deliver impact. You thrive in ambiguous environments with a low-ego mindset, leveraging AI coding tools as core collaborative partners in your engineering workflow.

• Lead product discovery alongside Product Managers and domain experts to understand real workflows, map them out, and define clear problems, baselines, and success criteria as the foundation for building agentic solutions.
• Design, build, test, and deploy production agents, automations, and internal tools using coding agents, LLM APIs, and agent frameworks, integrated with Rubrik’s internal agent platform and business systems.
• Turn complex processes into multi-agent workflows that carry real work and make decisions: orchestrators that qualify and routes deals, workers that move orders through approval and fulfillment under business rules, agents that monitor signals and recommend actions for customer health, or coordinated agents that handle hire-to-retire steps across HR, IT, and managers with shared state and clear handoffs.
• Engineer agents for autonomy with reliability, including tool and orchestration flows, context management, caching, retries, and failure handling.
• Leverage full-stack engineering across backend services, APIs, data models, interactive screens, approval UIs, and chat interfaces as the complete medium to deliver high-ownership, user-facing agent systems.
• Apply solid software engineering practices to AI systems, treating prompts, skills, configs, and specs as versioned artifacts, with code review, tests, CI or CD, and staged rollouts.
• Implement evaluation, guardrails, and quality control: derive evals from workflow data, establish quality gates and guardrails, integrate human-in-the-loop approvals where needed, and monitor production runs to refine agents over time.
• Use measured results to decide how to evolve or retire agents, and turn successful systems into patterns and libraries that can be reused.

• Around 7 or more years of professional software engineering experience in production systems, and around 2 or more years of professional, hands-on experience building AI agents and agent powered production applications with measurable outcomes.
• Strong full stack engineering background, with a track record of designing, building, and owning services end to end, including backend APIs and integrations, data models, and user facing interfaces such as approval UIs, dashboards, or chat surfaces.
• A builder mindset: comfortable defining baselines and metrics, reporting results in business terms, working in a fast moving environment, and closing the loop by building, demoing, learning, and building again. If you like seeing improvements in hours saved, errors avoided, and teams unblocked, you will have plenty of chances to measure that here.

• System design: break down complex problems, design APIs and data models, reason about latency, throughput, and failure modes, and choose simple, robust architectures.
• Algorithms, data structures, access control permissions, and data provenance.
• Spec driven development: turn vague requirements into clear problem statements, write or refine design docs, and use a Spec → Plan → Task → Code→ Converge flow so work is traceable from design to code, including when you use AI coding agents.
• Execution discipline: use version control, code review, automated tests, CI or CD, staged rollouts and rollbacks, logging, metrics, and tracing, and improve systems based on issues and signals from production.

• Build agents in code using LLM APIs, SDKs, and libraries, not just no code or low code tools, and have shipped agentic systems that real users rely on.
• Design robust agent harnesses and multi agent orchestrations: model tools and resources, manage context and state, and keep runs reliable under real world data and load.
• Establish continuous evaluation and safety mechanisms: turn production failures into test cases, gate changes on eval suite results, and enforce guardrails across agent workflows.
• Exercise clear judgment on AI fit: decide when a problem needs an agent versus a simpler automation or no build, and make tradeoffs around model choice, latency and cost, context strategy, and autonomy using data rather than intuition.

• Knowledge/context graphs: entities and relationships, GraphRAG, hybrid retrieval.
• Agent frameworks and runtimes: Google ADK, LangGraph, OpenAI style, cloud or container platforms.
• Backend: Python, FastAPI or similar stacks.
• Tools integration: MCP, Slack or Teams bots, REST APIs, webhooks.
• Data systems: Snowflake, Salesforce, graph stores.
• UX: conversational flows, chat surfaces, guided forms, simple dashboards for agent output.