question regarding translation quality assurance.

Hi, I have a question regarding translation quality assurance.

I recently translated an entire technical site (about 180 pages, 500,000 Japanese characters) using the DeepL API through the DeepL MCP Server.

While the results were excellent, I still rely on a review workflow to ensure correctness, especially because I’m not a native English speaker.

Currently, my QA flow is:

1. Translate with DeepL API

2. Perform a round-trip check (JA → EN → JA)

3. Use GPT/Claude for final polishing

4. Validate terminology consistency manually

The specific flow is below (in Japanese?)

https://github.com/shuji-bonji/Notes-on-natural-language-translation/blob/main/translation-QA-process.md

My questions are:

- Are there any official or recommended best practices from DeepL for QA of large-scale technical translations?

- Does DeepL plan to offer any features such as real-time translation quality scoring, consistency checks, terminology validation, or round-trip verification in the future?

- Are there existing tools or workflows recommended by DeepL for ensuring translation correctness when working at this scale?

Thanks in advance — and thank you for the amazing tools!