Inferact jobs, judged for you, free

What they do

Inferact sells a platform that runs AI models in production. This work is called inference. Customers run the platform in their own cloud, or on GPUs that Inferact provides. The founders created vLLM, an open-source inference engine, and the team still leads its development. Inferact says that it serves more than 7 trillion tokens a day.

What they value

• Build in the open: its work on vLLM goes back to the community.
• Research is the product: research becomes faster, cheaper inference.
• Grow with the ecosystem of model labs, chip makers, and contributors.
• High ownership and clear technical communication.
• A close focus on benchmarks and speed gains.

Where they hire

Most roles are in San Francisco, the headquarters. Seven are in Singapore, the hub for Asia and the Pacific. Three roles are remote, worldwide, with regular overlap with Pacific Time. Thirteen San Francisco roles will consider remote work in the US for exceptional people. Other roles are in the office. Most postings say that Inferact sponsors visas case by case.

Teams and roles

Research and Engineering has 24 of the 33 roles. Its engineers have the title Member of Technical Staff.

• Kernels: three kernel roles, for NVIDIA GPUs, AMD GPUs, and Google TPUs, each in San Francisco and Singapore. A kernel is low-level code that runs on the chip.
• Inference and scale: engineers who make the vLLM engine faster, run it across thousands of accelerators, and manage the clusters and deployments.
• Security and reliability: product security, vulnerability management, and site reliability for production.
• Customers and developers: a forward deployed engineer who works inside customer teams, and a developer relations role.

Other roles are in marketing, sales and partnerships, finance, and IT. A co-op role is for University of Waterloo students. Inferact raised $150M at a value of $800M in January 2026, in a round that a16z and Lightspeed led. Most engineering roles pay $200K to $400K in San Francisco, or S$200K to S$400K in Singapore, plus equity.

What they look for

Most engineering roles ask for a bachelor's degree or equal experience. More than half ask for Python. About half ask for CUDA or Kubernetes. Kernel roles ask for deep work on GPU or TPU kernels, and for profiling tools. Many roles prefer experience with vLLM or another inference engine, and contributions to open-source projects.

The two product marketing roles ask for experience in marketing to developers or developer tools. One of them asks for regular use of AI tools.