Box

Software Engineer III, Agents Platform

Redwood City, CA, United StatesFrom $206.5kmidAdded today

About this role

Box is hiring a Software Engineer III to build core components of an enterprise-grade AI Agents Platform built on LangGraph. You'll design scalable, secure services that enable teams to build and deploy AI agents for real-world workflows, focusing on multi-tenant architecture, orchestration, and developer tooling.

What you'll do

Design and build core Agents Platform components that power agentic use cases like Deep Search and Deep Research
Develop agent repositories, SDKs, and developer tooling to streamline agent development, testing, and deployment
Build multi-tenant control plane services enforcing isolation, resource allocation, and per-tenant SLAs
Create high-throughput, low-latency services with strong observability, guardrails, and secure execution
Collaborate with ML engineers and product teams to translate requirements into LangGraph-based capabilities
Participate in on-call rotations, production support, and root cause analysis for reliability improvements

What they're looking for

Backend or platform service development at scale (3+ years)
Object-oriented languages (Java, C++, C#, Scala)
Distributed systems and platform architecture
API design and microservices patterns
Production observability (logging, metrics, tracing, alerting)
LangGraph or agent framework experience (nice to have)
LLM concepts and cloud ML platform integration (nice to have)
Kubernetes or containerized workload management (nice to have)

Benefits

Opportunity to work on AI-first platform transforming enterprise workflows
Collaborate with ML engineers and cross-functional teams
Minimum 3 days per week in-office at Redwood City headquarters
Part of Box's mission to bring intelligence to content management

Likely interview questions

Describe your experience building and operating backend services at scale—what were the key challenges around reliability and observability?
Tell us about a time you designed a multi-tenant system. How did you approach isolation and fair resource allocation?

Have you worked with agent frameworks like LangGraph or LangChain? What patterns did you find most effective?
Walk us through how you'd approach building an SDK or developer platform. What makes it successful from a developer experience perspective?
Describe a production incident you debugged. How did you conduct root cause analysis and prevent recurrence?
How do you approach end-to-end project ownership—from scoping through rollout and iteration?
What's your experience integrating with multiple LLMs or ML platforms, and how have you handled the complexity?

Unlock all 7 questions free — and practice them live →

Apply on the employer's site →

Opens the official application on the employer’s site. No login required.