Sieve is a specialized cloud built for video / audio AI.

Every video product today is being overwhelmed by a ton of new use cases that are enabled by AI. Video is unique in that it’s much more compute and data-intensive to process or generate compared to other modalities. This leads to a ton of complexity around the ways models are run, the kinds of extra processing needed to happen around them, and the complexity of pipelines that solve the most valuable use cases in the modality. To this end, Sieve’s strongest and most immediate value proposition comes from being an AI toolkit that solves problems unique to video — unlike generic AI developer tools that might exist today.

Video is just the start however. Learn more about our long term vision here!

About us

Video is becoming the dominant digital medium of our time, making up 82% of internet traffic and powering applications in media, communications, VR, gaming, and robotics. Sieve is building a suite of developer tools to understand, edit, and search video — an in turn is powering a variety of applications across these industries.

We’re a small team based out of San Francisco and New York. In the last few months, we doubled our revenue, signed multiple 7-figure customers, scaled to thousands of GPUs, and closed a Series A from some of the best investors in the valley. Working at Sieve means joining a fast-growing AI infrastructure company, with many opportunities to grow alongside it. Our team brings together a diverse set of backgrounds from academic PhDs to quant engineers and self-taught tinkerers.

About the role

As an applied research engineer at Sieve, you’ll work with some of the latest VLMs and generative models to build systems for a variety of video understanding, editing, and search use cases. Often this involves working on ambiguous research problems and finding clever techniques to solve them. You’re likely a good fit if you’re comfortable working with models and squeezing every drop of performance out of them through clever pre/post-processing, pipelining, inference optimization, and fine-tuning.

As an early engineer, you’ll get to:

• work on low level containerization, networking, and distributed systems problems impacting thousands of GPU workloads at any given moment
• write and deploy code that touches users every single day
• work with the latest generative models
• make decisions on product experience and design
• interact directly with customers in calls, shared Slack channels, etc.

We primarily code in Go, Python, and Rust. Additionally, we heavily utilize Redis, Prometheus, Loki, Grafana, MongoDB, Pulumi, and Clickhouse. Our frontend is build with Next.js.