Senior Software Engineer, LLM Platform

The role

This role builds and operates the shared LLM control plane and infrastructure for Handshake's career marketplace and Handshake AI. You will own LiteLLM gateways, provider integrations, and capacity management while contributing to broader Data and ML platform systems.

What you will do

• Build and operate LiteLLM-based AI gateways and shared LLM clients.
• Own provider and model onboarding, routing, failover, rate limits, and capacity planning.
• Build self-service virtual-key, model-access, budget, and credential-management workflows.
• Establish SLOs, observability, cost attribution, and alerts for production LLM traffic.
• Safely qualify and roll out new models, providers, SDKs, and gateway configurations.
• Support hosted and self-hosted inference through a consistent platform interface.
• Partner with product, AI, and FDE teams to turn recurring delivery problems into paved-platform capabilities.
• Contribute to the broader Data and ML Platform, including workflow orchestration, model serving, shared cloud infrastructure, and developer tooling.
• Participate in team on-call and support, improving runbooks, automation, and reliability across owned platform services.

Required qualifications

• Strong production software engineering experience in Python, TypeScript, Go, or a similar language.
• Experience building or operating high-throughput API gateways, proxies, or multi-tenant platform services.
• Hands-on experience with Kubernetes, Terraform, CI/CD, and production service ownership.
• Practical experience with LLM provider APIs, including streaming, long-running requests, retries, timeouts, cancellation, and rate limits.
• Experience with authentication, quotas, credential management, tenant isolation, and auditability.
• Experience building observability for distributed systems and leading production incident response.
• Experience with usage metering, cost attribution, capacity planning, or FinOps.
• Experience with data or ML platform systems such as BigQuery, Airflow, streaming pipelines, model serving, or ML observability.
• Strong judgment in ambiguous environments and a bias toward simple, reliable systems that scale.

Nice to have

• LiteLLM, Portkey, or a comparable multi-provider AI gateway.
• vLLM, Modal, Ray, Triton, PyTorch, or GPU-backed serving.
• Temporal or another durable-execution system for long-running LLM requests.
• Evaluation, batch inference, fine-tuning, RL, or other post-training infrastructure.
• Agent runtimes, sandbox infrastructure, MCP, tool use, or coding-agent infrastructure.

Benefits

• Equity in a fast-growing company
• 401(k) match
• competitive compensation
• financial coaching
• Paid parental leave
• fertility benefits
• parental coaching
• Medical, dental, and vision
• mental health support
• $500 wellness stipend
• $2,000 learning stipend
• ongoing development

About the company

Handshake powers 25 million job seekers, 1 million+ employers, and 1,600 educational institutions. Handshake AI works directly with frontier AI lab researchers to create evaluations, publish benchmarks, and push the boundary of data.