Senior Content Engineer

Lingo.dev is the localization engineering platform: the best way to measure translation quality, translate with LLMs, and proofread with native speakers. Teams including Mistral, Solana, SoSafe, and Cal.com run their languages on it, and over 300M words have been processed on the platform.

Seven people, fully remote, SOC 2 Type II compliant, backed by $4.2M from Y Combinator, Initialized Capital, and the founders of Supabase and Dependabot.

We are small on purpose. Everyone here ships to production, talks to customers, and owns a surface end to end. There is no layer between the person who builds a thing and the person accountable for it working. If you want a narrow role with a queue of tickets handed to you, this is the wrong company. If you want the whole problem, this is the right one.

Lingo.dev is the localization engineering platform. Teams measure translation quality, translate with LLMs, and proofread with native speakers. We are a small YC-backed team, fully remote, with paying customers.

This is our first marketing hire. You own content end to end: creation, publishing, distribution, measurement, and automation.

Apply through our careers page, not through this listing: https://lingo.dev/en/careers/ff88132a-cb79-4d35-a6a0-a6230de7a013

• Blog posts, research articles, essays, changelogs, the newsletter
• Distribution on LinkedIn, dev.to, Hacker News and Reddit
• Interviewing our engineers and customers, then writing for both technical and non-technical readers
• Weekly publishing, with volume growing through Claude workflows and templated distribution
• SEO, GEO and AI search visibility, measured in Ahrefs, Search Console and the ad accounts
• The outcomes: signups, citations, rankings, backlinks

• You write plainly. No superlatives, no sentences that carry nothing.
• You have published something that worked: a newsletter with real subscribers, a post that traveled, technical content people shared.
• You code enough to automate your own workflow and distribution.
• You interview well and get people on the record saying something quotable.
• You use Claude daily as a production method and have opinions about prompting and output quality.
• You understand what Hacker News, the LinkedIn algorithm and AI systems reward.
• You work from Ahrefs and Search Console data.

Nice to have: marketing to skeptical audiences, published research or essays that changed how people think about something, video production.

You do not use Claude daily. You plan work from keyword spreadsheets. Your job ends at publishing. You need a weekly brief. You call yourself non-technical. Your portfolio is generic. You want to manage a team right away.

• Original engineering research to write about
• The whole pipeline, from draft to measured impact
• Claude Code and Claude Cowork budget for volume production
• Small team, direct founder access
• Conference and learning budget, flexible hours, health insurance
• Occasional in-person team meetings

Apply on our careers page: https://lingo.dev/en/careers/ff88132a-cb79-4d35-a6a0-a6230de7a013

The form there asks a few opinion questions instead of a CV. That form is the only application we review, so applying through this listing alone will not reach us.

The platform runs on Cloudflare: Workers, Durable Objects, and Workflows. Effect TS is the primary runtime. Transactional data sits in Postgres on Supabase. Analytics and event sourcing run on ClickHouse, with a custom Petri-net-inspired derivation engine over the event stream and a complex event processing layer modeled as Mealy machines. The frontend is TanStack Start.

The interesting problems are in the middle. A localization engine is a stateful translation API: it holds a team's glossary, brand voice, and per-locale model preferences, and every request retrieves the relevant slice of that context at inference time. Getting that retrieval right across dozens of locales, at volume, inside a latency budget, is the core engineering work.

Around it sits scoring, which has to produce a number a human will defend to their VP, and a correction loop that turns a native speaker's edit into a rule the engine applies to everything after it. We are model-agnostic by design: teams route per language pair, with fallback orders, on their own provider keys.

Four steps, and we move quickly.

1. Screening call with Max, our CEO, around 30 minutes. We talk about what you have published, which numbers moved because of you, and what you want to own. Compensation is discussed on this call.
2. Trial project. A real, small piece of work, followed by an interview where we go through your reasoning and your drafts together. This is the step that decides it, not the CV.
3. Offer.
4. Four-week probation period, so both sides find out whether this works in practice before either of us commits further.

You hear back from us within a few days at every step. If the answer is no, we tell you why.