Streaming Platform Technical Analysis
Budget: $10 – $30 USD
I need a seasoned reverse engineer who can take a deep dive into a particular video-streaming site and return with a clear, thorough explanation of how it actually works. While I want the full end-to-end picture, my main curiosity lies in the video-player layer—specifically how it performs AI-based upscaling during playback.
Here is what I expect you to uncover and document:
• Front-end stack in use and any frameworks that stand out.
• Detailed breakdown of the video player itself—controls, custom scripts, third-party libraries, and, most importantly, the AI upscaling workflow (models employed, in-browser processing vs. server-side inference, adaptive behaviour, fall-backs).
• Exact video delivery approach: HLS, DASH, WebRTC, peer-assist, or any hybrid they might have built.
• CDN path, edge logic, tokenised URLs, and the full streaming workflow from origin to player.
• All API calls the player makes, how sessions are authenticated, key headers, and data formats passed around.
• Caching and performance tricks you observe—service workers, IndexedDB, pre-fetching, or smart chunking.
• Any other standout features (e.g., caption handling, multi-track audio, ABR logic) and how they slot into the architecture.
• A high-level diagram that ties these pieces together, showing front end, backend services, CDNs, AI components, and data flow.
Deliverables
1. Written technical report (PDF or Markdown) that walks through each item above, citing request/response examples, code snippets, or packet traces where helpful.
2. An architecture diagram (drawn in any clear tool—Lucidchart, draw.io, or similar).
I will give you the target URL once we start. Passive observation and tooling such as developer tools, Wireshark, Charles Proxy, or custom scripts are all fine; absolutely no disruptive or intrusive testing. The objective is insight, not imitation.
If you have solid experience dissecting streaming stacks and can articulate complex systems in plain language, I’m ready to work with you.
Here is what I expect you to uncover and document:
• Front-end stack in use and any frameworks that stand out.
• Detailed breakdown of the video player itself—controls, custom scripts, third-party libraries, and, most importantly, the AI upscaling workflow (models employed, in-browser processing vs. server-side inference, adaptive behaviour, fall-backs).
• Exact video delivery approach: HLS, DASH, WebRTC, peer-assist, or any hybrid they might have built.
• CDN path, edge logic, tokenised URLs, and the full streaming workflow from origin to player.
• All API calls the player makes, how sessions are authenticated, key headers, and data formats passed around.
• Caching and performance tricks you observe—service workers, IndexedDB, pre-fetching, or smart chunking.
• Any other standout features (e.g., caption handling, multi-track audio, ABR logic) and how they slot into the architecture.
• A high-level diagram that ties these pieces together, showing front end, backend services, CDNs, AI components, and data flow.
Deliverables
1. Written technical report (PDF or Markdown) that walks through each item above, citing request/response examples, code snippets, or packet traces where helpful.
2. An architecture diagram (drawn in any clear tool—Lucidchart, draw.io, or similar).
I will give you the target URL once we start. Passive observation and tooling such as developer tools, Wireshark, Charles Proxy, or custom scripts are all fine; absolutely no disruptive or intrusive testing. The objective is insight, not imitation.
If you have solid experience dissecting streaming stacks and can articulate complex systems in plain language, I’m ready to work with you.