Bitu Technology is the registered company of Tubi's China team. We are engineers passionate about leveraging advanced technology to improve lives, and we hope to use this channel to connect and advance together.
The article explains how Tubi, an ad‑supported streaming platform with tens of millions of users, migrated from a Node.js‑Spark‑Redis stack to a Scala‑based backend to achieve sub‑10‑millisecond recommendation latency, improve fault tolerance, and simplify domain modeling using Akka and functional programming.
This article explains how Tubi’s machine‑learning team reduces exposure bias in its video recommendation pipeline by normalizing popularity features, incorporating additional signals such as search behavior, and applying exploration techniques like bandit algorithms to diversify content exposure.
The March 19 Elixir Meetup brought together four seasoned Elixir practitioners and over 3,200 online attendees to discuss resilience with Elixir and Kubernetes, FaaS design using Arweave and LiveView, native compilation of Elixir, Zig‑based NIF development, and to share learning resources and community opportunities.
The article analyzes the rapid rise of ad‑supported streaming (AVOD) in the United States, highlighting Tubi's 2021 viewership surge, pandemic‑driven user behavior shifts, demographic trends, advertising revenue growth, and future content and hiring plans.
The talk explains how Tubi's Multimedia Processing Platform uses an interface‑driven architecture, Elixir protocols, and composable movers and comparators to migrate petabyte‑scale video files across cloud storage while balancing performance, cost, and testability, offering practical insights for Ruby and backend engineers.
The article details Tubi's Multimedia Processing Platform (TMPP), describing its architecture, processing stages, resource management, and distributed task scheduling for large‑scale video transcoding and delivery across multiple devices.
This article explains how Tubi leverages machine learning—particularly PyTorch, Databricks, and cloud services—to improve content understanding, advertising technology, and recommendation systems within its advertising‑based video‑on‑demand platform, outlining the three AVOD pillars, technical stack, and future research directions.
At Tubi, the Popper experiment engine—a Scala‑based, Akka‑powered backend service—combined with a self‑serve UI, automated analysis pipelines, and rigorous health checks, enables teams across ML, mobile, and OTT to run scalable A/B tests, rapidly iterate, and make data‑driven product decisions.
The 2021 Online Scala Meetup organized by Tubi featured four technical talks covering a fully reactive ad‑serving platform built with Scala and Akka‑Streams, a pragmatic take on functional programming, Scala 3 typeclass implementation, and a Spring‑Akka microservice integration, followed by summaries and recruitment information.
This article recounts how a Tubi engineering team diagnosed a surprising latency increase after adding Redis cache to an Elixir manifest‑fetch API, identified a hidden CPU‑heavy Regex.replace/4 bug, applied targeted optimizations, and contributed a fix to the Elixir core.
