arvida

Working on https://localhero.ai, on-brand automated translations for product teams. It mainly runs as a GitHub Action, translates new strings on your PRs against your glossary and style guide. This month focus went mainly on MCP and stuff around that. Thinking a PM/designer/etc who wants to bulk tweak copy without touching git or similar. Claude Desktop/ChatGPT turns out to be a pretty nice surface for this, the assistant finds every key that needs changing and organises the work, Localhero keeps it on-brand. A pretty sweet workflow. Also doubled down on Lingui and Django support. Also been fighting some spam. A 400 from our mail service showed up in error tracking, looked like a malformed address, turned out something else. Someone setup a bunch of bots to signup and try so send crypto spam using the team invite function. Kind of persistent, I fix a few things and it tries a few others. Finally stopped, interesting to see it live like this.

On the side, https://infrabase.ai (hand-verified AI infrastructure directory) launched a Media Generation category, image and video APIs after a run of requests for it. Also the submission spam is way up here latley, been improving the detection here as well. Also seeing real interest in paid listings now, including one that started as a link-buying request I turned down :)

Been doing a lot of AI assisted emailing, so I have started to work on a little review UI for AI generated text, comments on specific words, inline edits, Claude picks up both and revises in place. Like a PR review interface for text. Turns out pretty good workflow with Claude watching the files for new comments, editing and commenting back right away.

Working on https://localhero.ai, automating i18n translations for product teams (runs as a GitHub Action, translates new strings on PRs matching your brand voice and glossary).

Shipped "bring your own translations" feature this month, built together with a customer who wanted Localhero for the workflow and review layer but writes some translations themselves. Otherwise smaller improvements and marketing. Better Lingui support, for example: projects using explicit IDs keep the source text in the source locale's msgstr instead of the msgid, which our CLI got wrong before. Also wrote some blog posts, a comparison of GitHub Actions for i18n and a guide to auto-translating gettext .po files in CI (Django, Phoenix, LinguiJS). Content is slow going but it seems to be how people find us via perplixity/gemini/chatgpt.

On the side, https://infrabase.ai (AI infrastructure tools directory) grew ~16% MoM. I leaned further into the agent angle: added a public query API plus llms.txt/agents.md/etc so agents can query the directory directly, and exported the full dataset to a public GitHub repo. Starting to get som traction and people wanting collaborations.

Working on https://localhero.ai, automating i18n translations for product teams (runs as a GitHub Action, translates new strings on PRs matching your brand voice and glossary, skills for Claude etc). A few new paying customers this month. One is an app I've used myself for years to find hiking trails in Sweden, which was a fun email to get (Elixir/Phoenix with gettext/.po).

One feature I have been working on this month is "bring your own translations". Been talking with a few customers about how they work with Claude Code and similar and one wants Localhero for the process and the review/consistency layer on some of their languages, not the AI. So I've been building a better flow for mixing manual translations with AI generated ones, together with them.

On the side, https://infrabase.ai (AI infrastructure tools directory) had its best month again, up ~55% MoM unique visitors. Still chasing the GEO angle, visitors from ChatGPT/Perplexity/Gemini average 2-3 minutes on the site vs ~1 minute for direct traffic. People are starting to find it on their own too. Had quite a few ask to get their products listed and a couple of partnership conversations.

Working mostly on https://localhero.ai, automating i18n translations for product teams (runs as a GitHub Action, translates new strings on PRs matching your brand voice and glossary). Got a new paying customer just now, a pretty big Swedish company, been working closely with the team to add an improved review flow to match their needs.

Most interesting work this month came from a customer in Japan who wanted something I hadn't built called "Easy Japanese" (やさしい日本語), which is like a simplified version for non-native readers. Turned into supporting fully custom locales. Also been improving how the GH action picks up manual changes to target translations in commits, they now get ingested back for review, making things a bit more flexible.

On the side, https://infrabase.ai (AI infrastructure tools directory) had its best month yet, up ~40% visitors MoM. Interesting finding is still the GEO angle, visitors arriving from Claude/ChatGPT/Perplexity spend 2-6 minutes on the site. Been making comparison pages more "extractable" for AI answers with job-tagged tables and some have now held their Perplexity citations for weeks.

Good start of the month with user discussions and feedback.

Mainly working on https://localhero.ai, automating i18n translations for product teams. Basically runs as a GitHub Action, translating new strings on PRs matching your brand voice and glossary. Got our first fully selfserve customer a few weeks back (found us through the docs). Interesting work lately has been improving how the system learns from manual edits, when someone tweaks a translation in the UI, it feeds back into translation memory and influences future translations in a smart way. Also did stuff like improving our agent skill, so coding agents get glossary/style guide context automatically and they can write source copy that better matches the brand.

Been pushing some new stuff on https://infrabase.ai as well, my AI infrastructure tools directory. Traffic growing steadily from comparison and alternatives pages. Interesting finding is that blog posts rank better but get fewer clicks now because AI Overviews, interactive comparison pages still earn clicks. ChatGPT has also started citing the site more as a source. Adding new content and polishing existing parts of it, added a page focusing on EU based services at https://infrabase.ai/european.

Been working on https://localhero.ai, its my service to automate on-brand translations for product teams. I've been doing outreach to Swedish companies/people, getting some good interest from a few that want to automate their localization workflow but don't want the work of maintaining own solutions. Even though you can build a version working with coding agents these days, there is a lot of stuff around it to make it work well over time in a product org. On the tech side for Localhero, one thing I've been working on how it learns from manual edits. Like when a PM or designer tweaks copy in the Localhero UI, those things now better feed back into a translation memory and influence future translations. It's like a self-learning loop, turns out a pretty nice combo of using old-school techniques and offloading some work to LLMs.

Also been spending some time on my old side project https://infrabase.ai, an directory of AI infra related tools. Redesigned the landscape page (https://infrabase.ai/landscape), going through product submissions and content, optimizing a bit for seo/geo.

Cool, I'll drop a dm on linkedin!

Building https://localhero.ai, automated on-brand i18n translations that run in your CI pipeline. Right now I'm working on better .po/gettext support, based on feedback from an early customer. With gettext you usually keeps source strings in the actual source code. So I'm building a workflow where non-technical people (PMs, designers) can edit translations in the web UI and then easily generate a PR with both code changes and translation file updates. Trying to make translations work smooth for both automated CI pipelines and PMs/designers who don't live in Git, when translations are checked into the repo. Also going through my network, talking to devs and localization folks to understand what could be improved in their orgs for translations.

Building https://localhero.ai, automated on-brand i18n translations that run in your CI pipeline.

Right now I'm working on better .po/gettext support, based on feedback from an early customer. With gettext you usually keeps source strings in the actual source code. So I'm building a workflow where non-technical people (PMs, designers) can edit translations in the web UI and then easily generate a PR with both code changes and translation file updates. Trying to make translations work smooth for both automated CI pipelines and PMs/designers who don't live in Git, when translations are checked into the repo. Also going through my network, talking to devs and localization folks to understand what could be improved in their orgs for translations.

I'm building https://localhero.ai. Automated i18n translations that run in your CI pipeline.

I have worked as dev in many different constellations of the years, and seen many teams choose between bad options like delay feature launches for manual translations, ship incomplete translations and promise "we'll translate it later," or lately use ChatGPT/LLMs that lose consistency/context and require coordination.

Localhero starts from the premise that translations are part of CI. New strings get translated automatically in GitHub Actions, with glossaries and style guides so it sounds like your product and not generic AI output.

Goal is to help product teams ship localized features without all the coordination/delay.