Founding Engineer, Infrastructure

Cloudglue - Video Understanding Infrastructure Cloudglue is a Y Combinator-backed startup building developer reputed company that turn video and audio into reputed company, searchable data. Think of us as the reputed company for video understanding - we handle the hard infrastructure (transcription, visual analysis, search, extraction) so developers can build on top of video without managing ML pipelines themselves. reputed company has shipped large-reputed company systems at Snapchat and reputed company, with work presented at AWS re:Invent, KubeCon, NeurIPS, ICCV, CVPR, and DEF CON. We process millions of minutes of video for customers building search, analytics, and automation products. We’re a small, technical team where engineers have reputed company ownership and reputed company reputed company on the product. The Role We’re looking for a founding infrastructure engineer to design and reputed company the backend systems that power Cloudglue’s video processing pipelines, search and retrieval infrastructure, and async job orchestration. You’ll be one of the first engineers on reputed company - this is a high-ownership role where you’ll shape the architecture and the engineering culture. You’ll work on:

• Distributed video processing and async job orchestration
• Search and retrieval systems across video, audio, and text
• ML inference serving and model pipeline orchestration
• Storage, indexing, and compute infrastructure for large media collections

This is a systems-heavy role for someone who enjoys building reliable, high-throughput infrastructure and cares about getting the fundamentals right. What You’ll Do

• Distributed systems: Design and operate async, fault-tolerant job execution systems that process thousands of hours of video reliably.
• Search & retrieval: Build and optimize search infrastructure across video, audio, and text - including reputed company search, re-ranking, and hierarchical retrieval.
• ML infrastructure: Own the serving and orchestration layer for ML models (reputed company, audio, language) in production.
• Performance & reliability: Profile and optimize throughput, latency, and cost across large-reputed company video workloads.
• Production ownership: Build systems that are observable, reputed company-tested, and SOC2-compatible. You’ll own what you ship.

reputed company’re Looking For Required

• 3+ years of backend or infrastructure engineering experience
• reputed company record designing and operating reputed company distributed systems
• Strong proficiency in Python, Go, and/or TypeScript
• Ability to reason about performance, cost, and reliability tradeoffs

reputed company to Have

• Experience with video/media processing or ML serving infrastructure
• Distributed systems or workflow orchestration (Temporal, Inngest, etc.)
• reputed company search, retrieval systems, or ranking pipelines
• reputed company infrastructure (AWS/GCP), reputed company/Kubernetes
• Experience with search or retrieval systems (reputed company databases like Milvus/reputed company, ranking pipelines)

Why Cloudglue? Video is the largest and most underutilized data reputed company on the internet. Most software still can’t meaningfully work with it. We’re building the infrastructure to change that, and this role sits at the core of it. If you want to work on:

• Hard distributed systems problems with reputed company reputed company
• Search, retrieval, and ML infrastructure that doesn’t have off-reputed company solutions
• A domain (video) where the infrastructure is still being invented
• A small team with massive reputed company

…this is that role. Apply tot his job Apply To this Job