Claude-ts – Translation proxy to fix non-English token waste in Claude
Fixes multilingual token waste by translating to English before Claude, not after.

This isn't a startup — it's archival work: a first-time English translation of an obscure Einstein paper with corrections folded into the text and an institutional DOI, which makes it immediately citable. The landing page is functional and trustworthy (Cambridge Apollo hosting), but there's an odd mismatch on rights metadata — the author claims public domain timing while the record shows 'All Rights Reserved' — which deserves a quick fix for clarity.
Historians of science, physicists, physics students, educators, and science history enthusiasts
I have a blog post that gives a little more context: https://adaptive-machine-patterns.com/blog.html#einstein [alt. link: http://archive.today/381Pl] I am new to blogging, so advice welcome.
The preprint is hosted at CULA repository (many thanks) https://www.repository.cam.ac.uk/handle/1810/398349 and it has a DOI: https://doi.org/10.17863/CAM.127224
Fixes multilingual token waste by translating to English before Claude, not after.
Turns impenetrable weather jargon into readable English, covering 19 NWS offices.
Translation proxy for one forum—useful but technically straightforward.
Translation of existing Chinese Python course—useful but nothing technically new here.
Automated PubMed summaries emailed weekly for eight specific research topics.
Persona-based tone switching captures cultural nuance better than Google Translate ever could.