Data scientist - Agentic AI

Description

This is an onsite role, with most of your time spent in a company office.

The company builds edge-to-cloud technology that helps organizations connect, secure, analyze and act on their data and applications, wherever they run.

You will build and run the core agentic workflows behind a next-generation AI assistant for network operations.

The work sits where data science, generative AI and production engineering meet. You will design, build and evaluate the reasoning pipelines, tool-calling patterns, skills and MCP server integrations that let the assistant diagnose, troubleshoot and resolve complex networking problems on its own.

The ideal person has solid hands-on experience with LLM agent frameworks, LangGraph especially, and the engineering discipline to ship reliable, observable AI systems in a cloud-native setting.

This is a subject matter expert role in agentic AI and applied GenAI. Your work shapes technical components of products, solutions or services on an ongoing basis. You will advise project teams, often across functions, decide independently how to reach goals, and may lead and mentor others.

Responsibilities

• Build and refine agentic workflows in LangGraph, covering ReACT orchestration loops, dynamic tool selection and binding, multi-step reasoning and self-correction.
• Create and maintain MCP servers and skills: define tool schemas, implement domain tools, write SKILL.md playbooks, and manage versioning, deployment and monitoring.
• Make LLM capabilities production-ready, including structured outputs, streaming, tool calling, prompt design and strong error handling across agent execution paths.
• Develop retrieval and memory services: RAG pipelines, vector-backed semantic search, hybrid retrieval, long-term agent memory (semantic, episodic, procedural) and relevance tuning.
• Design and run evaluation frameworks for non-deterministic agent systems: set metrics, build test harnesses, run A/B tests on skills and tool setups, and keep raising quality.
• Work with network engineers and product managers to turn troubleshooting playbooks into agentic workflows built from skills, tools and data pipelines.
• Build data analysis and transformation code that runs in sandboxed execution environments (Code Mode), including multi-tool orchestration scripts, aggregation and visualization.
• Deploy and operate containerized services on Kubernetes, and contribute to CI/CD pipelines, container images, health probes and resource tuning.
• Own observability for agentic workflows: tracing, logging, cost tracking and performance monitoring to keep non-deterministic systems reliable in production.

Requirements

• Master's or PhD in computer science, data science, mathematics, statistics or a closely related quantitative field.
• Typically 4, 6 years building production ML or AI systems, including 1, 2 years of hands-on work with generative AI and LLM applications.
• Production experience with agent orchestration frameworks beyond prototypes, such as LangGraph (strongly preferred), LangChain, Claude Agent SDK or equivalent.
• Working knowledge of agent design patterns: ReACT loops, tool and function calling, dynamic tool binding, skill-based execution, multi-step planning and self-correction.
• Hands-on work with MCP or similar tool-serving protocols, including tool schema design, server implementation and registry management.
• LLM API integration at scale, covering prompt engineering, structured outputs, streaming, error handling and cost optimization.
• RAG pipeline design, including chunking, re-ranking, hybrid search, vector stores such as OpenSearch and relevance optimization.
• Experience building evaluation and testing frameworks for non-deterministic AI, such as offline evals, A/B testing and LLM-as-judge.
• Strong grounding in statistics and machine learning, such as anomaly detection, time-series analysis, clustering or causal inference.
• Applied ML judgment: knowing when to use retrieval versus fine-tuning, prompting versus structured generation, and how to debug model behavior in production.
• Proficient Python developer who has worked in production codebases, not only notebooks.
• Kubernetes experience deploying, scaling and managing workloads, including Deployments, Services, ConfigMaps, Secrets and health probes.
• CI/CD pipeline work for automated build, test and deploy, using Jenkins, GitHub Actions, ArgoCD or similar.
• Container image management with Docker, including building, tagging and versioning, plus familiarity with a registry such as ECR or GCR.
• Backend service development with FastAPI or equivalent, and REST or GraphQL API design.
• Observability for AI systems using tracing, monitoring and logging tools such as LangFuse or Prometheus.
• Preferred: experience with agent memory systems, such as LangMem or custom memory architectures.
• Preferred: familiarity with sandboxed code execution environments such as E2B or Firecracker.
• Preferred: networking knowledge, including wireless or wired diagnostics and troubleshooting. A strong plus but not required.
• Preferred: experience with AWS Bedrock, OpenSearch Serverless or similar managed AI/ML services.
• Preferred: strong written and verbal communication, with the ability to explain technical designs to senior leadership.

Conditions

Onsite role, primarily working from a company office.

Base salary range in the US is USD 155,500 to 315,000 per year in California. The actual offer may vary by location, experience, education or skill level.

Job level TCP04 in the Engineering function.

The company is an equal opportunity employer that encourages applications from people with disabilities, and accommodation requests can be made through its secure form.

Benefits

• Comprehensive health and wellbeing benefits covering physical, financial and emotional wellbeing.
• Career development programs, including options to move into another division.
• Flexibility to manage work and personal needs.
• Variable incentives may be offered on top of base salary.
• US benefit details are published on the company's new-hire enrollment page.