✦ Luna Orbit — Software Engineering

Software Engineer, Model Routing & Inference

at Cursor

📍 New York Unknown Posted April 07, 2026
Type Not Specified
Experience mid
Exp. Years Not specified
Education Not specified
Category Software Engineering

Cursor is seeking a Software Engineer to join the Model Routing & Inference team. You will build the inference platform that powers every AI interaction in the product, including the inference gateway and cross-provider routing and failover.

  • Build and maintain the inference platform
  • Design and evolve the inference gateway
  • Implement cross-provider failover
  • Design routing backpressure and admission control
  • Optimize GPU utilization and cost-performance at scale

The role focuses on building high-throughput, low-latency distributed systems for inference serving, with emphasis on gateway abstractions over provider APIs, routing logic, backpressure, and capacity planning to optimize GPU usage and cost/performance.

The ideal candidate is a mid-to-senior software engineer with experience building high-throughput, low-latency distributed systems for inference serving. They should excel at making trade-offs between reliability, cost, latency, and user experience in production-scale AI workflows.

high-throughputlow-latencyinference servinginference gatewaycross-provider failoverroutingbackpressureadmission controlGPU utilizationcapacity planning
experience with multi-provider inferencecost/performance optimizationproduction-grade systems
high-throughputlow-latencydistributed systemsinference servinginference gatewaycross-provider failoverroutingbackpressureadmission controlGPU utilizationcapacity planningproduction systemsscalability
high-throughputlow-latencydistributed systemsinference servinginference gatewaycross-provider failoverroutingbackpressureadmission controlGPU utilizationcapacity planningproduction systemsscalabilityGraphics Processing Unitprovider APIs
problem-solvingcommunicationteam collaborationownershipability to weigh reliabilitycostlatencyand user experience
Industry SaaS
Job Function Design and operate the inference platform powering Cursor's AI interactions.
Role Subtype Platform Engineer
software engineerinference platforminference gatewaycross-provider failoverroutingbackpressureadmission controlhigh-throughputlow-latencydistributed systemsinference servingGPU utilizationcapacity planningproduction systemsscalabilitycursornew yorkprovider APIsproduction-gradeconfig managementGraphics Processing Unit

Bachelor's degree or equivalent in IT or related field, Lack of experience with high-throughput distributed systems

Apply for this Position →

Get matched to jobs like this

Luna finds roles that fit your skills and career goals — no endless scrolling required.

Create a Free Profile