Backend Engineer – Inference Optimization at Vercel

Seattle, Washington, United States

Apply Now

$150,000 – $250,000Compensation

Senior (5 to 8 years)Experience Level

Full TimeJob Type

UnknownVisa

AI, Machine Learning, TechnologyIndustries

Requirements

Must have:
Deep experience in optimizing model inference pipelines, model quantization and KV caching
Proficiency in backend systems and high-performance programming (Python, C++, or Rust)
Familiarity with distributed serving, GPU acceleration, and large-scale systems
Ability to debug complex performance issues across model, runtime, and hardware layers
Comfort working in fast-moving environments with ambitious technical goals
Nice to have:
Hands-on experience with vLLM or similar inference frameworks
Background in GPU kernel optimization (CUDA, Triton, ROCm)
Experience scaling inference across multi-node or heterogeneous clusters
Prior work in model compilation (e.g., TensorRT, TVM, ONNX Runtime)
Hands-on experience with model quantization

Responsibilities

Own the design and optimization of inference pipelines for large-scale models
Work closely with researchers and infrastructure engineers to identify bottlenecks
Implement advanced techniques like quantization and KV caching
Deploy high-performance serving systems in production

Skills

Python

C++

Rust

model quantization

KV caching

distributed serving

GPU acceleration

vLLM

CUDA

Triton

ROCm

Vercel

Platform for building and deploying web applications

About Vercel

Vercel provides a platform for developers and businesses to build, deploy, and manage modern web applications. Its services include tools that enhance image and video workflows using AI features like smart cropping and object detection. Vercel simplifies the complexities of serverless architecture, allowing for global content delivery without extra infrastructure. The company ensures high security and uptime with features such as automatic HTTPS and DDoS protection. Unlike competitors, Vercel focuses on a managed global rendering layer and offers a subscription-based model tailored to various client needs, from individual developers to large enterprises. The goal of Vercel is to empower developers to create efficient and secure web applications.

San Francisco, CaliforniaHeadquarters

2015Year Founded

$547.6MTotal Funding

SERIES_ECompany Stage

Consumer Software, Enterprise Software, AI & Machine LearningIndustries

501-1,000Employees

Benefits

Health Insurance

Stock Options

Company Equity

Professional Development Budget

Unlimited Paid Time Off

Remote Work Options

Home Office Stipend

Risks

Increased competition in the cloud application platform space threatens Vercel's market share.

Rapid AI evolution may outpace Vercel's current offerings, risking competitive edge loss.

Reliance on a subscription model could be risky during economic downturns.

Differentiation

Vercel offers a managed global rendering layer for modern web applications.

The company provides advanced AI-powered tools for image and video optimization.

Vercel's platform supports full lifecycle media management with auto-tagging and access control.

Upsides

Vercel secured $250 million in Series E funding for growth and platform development.

The introduction of V0 enhances Vercel's offerings in AI-driven web development.

Recognition as a Visionary in Gartner's Magic Quadrant boosts Vercel's market position.

Land your dream remote job 3x faster with AI

Try Jobo Free