Provides introductions and tutorials on Bilibili-related technologies.
The article introduces FineVD, the first large‑scale multi‑dimensional UGC video quality dataset, and presents FineVQ, a unified model that predicts quality scores, attributes, and distortion types across six dimensions, achieving state‑of‑the‑art performance on multiple benchmarks and cross‑dataset evaluations.
The article outlines a comprehensive live‑streaming ecosystem governance framework that combines data‑mining engines, tagging platforms, rule‑based disposal mechanisms, and multi‑stage user touchpoints to improve content quality, compliance, and platform sustainability.
This article explains the design of an AI gateway that centralizes LLM access, detailing its background, overall architecture, authentication, quota management, multi‑model routing, load‑balancing strategies, multi‑tenant isolation, observability features, and the supported API protocols for enterprise integration.
Bilibili's R&D built a cloud-based broadcast console for the 2024 CCTV Spring Festival Gala, delivering 4K HDR streaming, AI SDR-to-HDR conversion, low latency, bandwidth‑aware transcoding, and a synchronized “send bullet screen” interactive feature using custom SEI timestamps for hundreds of millions of viewers.
Bilibili Comics deploys a three‑layer protection system—stereoscopic tracing, crawler defense, and encrypted channels—that dramatically reduces illegal distribution, boosts creator engagement, and creates a healthier comic ecosystem through real‑time monitoring, automated takedowns, and secure CDN delivery.
Bilibili performed client-side DCDN disaster-recovery drills using a self-built HTTPDNS to simulate DNS, CDN, and SSL faults; automated scripts across Android, iOS, and Web injected errors, measured rendering latency, validated immediate downgrade to commercial services, refined fallback strategies, and demonstrated near-zero user impact during a real network incident.
To meet Bilibili’s rapidly expanding game business, the team built a unified real-time data warehouse using Hologres and Flink that replaces the traditional Lambda stack, delivering high-throughput writes, low-latency processing, seamless offline-online integration, global deployment, and real-time support for operations, advertising, and risk analytics.
By replacing the legacy FLV recorder with an m3u8‑based service, introducing a short‑segment, state‑machine transcoder, and deploying an event‑driven proxy and instant‑clipping UI, the team cut the 4‑hour‑40‑minute Spring Festival Gala’s live‑to‑VOD processing from 41 minutes to about eight minutes, achieving roughly a five‑fold speedup.
Bilibili’s engineering team built a scenario‑metadata and one‑click fault‑drill platform, implemented multi‑tier degradation, dynamic capacity planning, and extensive automated fault‑injection testing to guarantee zero‑severity incidents during the high‑traffic 2025 Spring Festival Gala live stream.
Bilibili’s ticketing platform evolved from a single‑transaction, synchronous design to an asynchronous batch system and finally to a Redis‑cached inventory layer, adding DB isolation changes, sharding, bloom‑filter protection and adaptive rate limiting, which together enabled handling up to 930 k requests per second and stable high‑concurrency ticket sales.
