Engineering Manager, Model Serving

Together AI•San Francisco, CA

12d•$250,000 - $300,000

About The Position

Together AI is building the AI Inference & Model Shaping Platform that brings the most advanced generative AI models to the world. Our platform powers multi-tenant server-less workloads and dedicated endpoints, enabling developers, enterprises, and researchers to harness the latest LLMs, multimodal models, image, audio, video, and reasoning models at scale. We are looking for an exceptional Engineering Lead to partner closely with our cross-functional engineering, infrastructure, research, and sales teams to ensure excellence of our ML API offerings. Your primary focus will be on delivering world-class inference and fine-tuning in our public APIs and customer deployments by building automation and operations processes. This role is ideal for a highly motivated and technically adept individual who excels in fast-paced, dynamic environments. You will be in charge of designing and scaling our ML processes & tooling at production scale – optimizing operations to ensure availability and reliability for our services, across differing tenants and user loads, and in a multi-cluster deployment. You will serve as a passionate advocate for internal and external customers, providing feedback to the wider engineering and infrastructure teams to improve our systems and core business metrics. If you thrive in a collaborative, problem-solving environment and are driven to deliver operational excellence, we encourage you to apply for this exciting opportunity.

Requirements

5+ years operating production ML inference or training systems at scale
2+ years in senior IC or tech lead roles, with demonstrated mentorship and technical leadership experience. Having built or scaled teams is a plus.
Deep expertise with Kubernetes, multi-cluster orchestration, and ML serving frameworks
Experience with multi-tenant SaaS platforms
Proven track record of SLA ownership with specific metrics (99.9% uptime, p99 latency targets)
Customer escalation and incident communication experience
Experience with LLM inference serving systems (SGLang, vLLM, TRT-LLM, or similar)
Ability to influence cross-functional teams and make deployment/architecture decisions

Nice To Haves

Experience building internal developer platforms or self-serve tooling
Background in cost optimization for GPU infrastructure
Contributions to open-source ML infrastructure projects

Responsibilities

Own availability and performance SLAs for production inference and fine-tuning services across serverless and dedicated deployments
Own & improve testing, deployment, configuration management, and monitoring practices for multi-cluster ML infrastructure – partnering closely with Infra SREs
Build self-serve tooling and automation to reduce operational toil and enable self-serve offerings.
Define and enforce configuration best practices for inference engines (SGLang, TRT-LLM, vLLM etc.) to prevent runtime issues
Lead incident response, conduct postmortems, and drive reliability improvements
Mentor team members and potentially grow into hiring/team building as the organization scales
Partner with infrastructure and ML engineering teams to improve system reliability and cost efficiency