Automating the Toil, Not the Community: AI-Assisted OpenTelemetry Docs Localization

Localization is a powerful, although often overlooked, gateway to open-source contribution. For projects like OpenTelemetry, translating documentation forces you to dive deep into the project, understand the core terminology, and navigate your first pull request.

It is a brilliant way to build confidence, until the friction hits. Managing formatting, context-switching, and OTel glossary alignment can easily turn a single-page translation into a repetitive 4-hour task. The cognitive toil is real, and it can cause localization efforts to stall.

What if we could automate the drag while preserving the human connection?

In this lightning talk, we will explore an AI-powered workflow that abstracts away the mechanical overhead of OpenTelemetry documentation localization while keeping the human in the loop.

We'll close with an open question for the ecosystem: If we lower the friction of contributing through automation, do we dilute the learning experience? Or can we actually make localization efforts sustainable, allowing new contributors to focus on learning about the project rather than fighting the linter?

Transcript

1. Automating the Toil, Not the Community AI-Assisted OpenTelemetry Docs Localization Imma Valls Staff Developer Advocate Grafana Labs
2. Agenda 2 01 Localization is contribution 02 How localization works 03 What is toil ? 04 AI-assisted localization 05 Automate the toil, not the community © 2026 Cloud Native Computing Foundation
3. Localization is contribution 3 © 2026 Cloud Native Computing Foundation
4. Why Localized Documentation Matters 🌐 🧠 🚀 Accessibility First More Than Translation Low-Friction Entry Ensures everyone can learn in the language they understand best. “Part of a project accessibility” A dual-learning mechanism to understand core concepts while mastering the contribution workflow. Learn concepts + workflow © 2026 Cloud Native Computing Foundation Smaller initial scope providing immediate usefulness without needing full codebase mastery. Ideal OSS onboarding
5. What you learn while translating 01 Learn the project 02 Learn to contribute Concepts, terminology, documentation, how things fit together GitHub, Pull requests, Reviews, Contribution workflow 03 Meet the people And meet the people behind the project 5 © 2026 Cloud Native Computing Foundation
6. Observability Core Vendor-Neutral Standard CNCF project dedicated to robust, modern cloud observability standards. APIs, SDKs, instrumentation, collector, semantic conventions, and more. Cloud-Native Ecosystem Documentation Widely adopted across a vast and rapidly growing cloud-native landscape. Documentation is a primary catalyst for user learning and project adoption. “A global project needs documentation that works for a global community” 6 © 2026 Cloud Native Computing Foundation
7. Supportive Community 💖 Documentation was my entrypoint Carol Valencia Pablo Baeyens Diana Todea Vitor Vasconcellos Fabrizio Ferri-Benedetti Severin Neumann … and many more amazing contributors! Welcoming, guiding, and empowering new contributors every day. © 2026 Cloud Native Computing Foundation
8. How localization works 8 © 2026 Cloud Native Computing Foundation
9. What gets localized English Source Docs The primary, canonical documentation for OpenTelemetry  Localized Versions Community-translated documentation across multiple active languages: Spanish · French · Japanese Korean · Polish · Portuguese Romanian … and more 9 © 2026 Cloud Native Computing Foundation
10. What does a localization contribution look like? Localization Lifecycle Workflow 01 02 03 04 05 06 Find Content Make Change Open PR Review Merge Live Docs Identify pages to translate or update Translate content ”locally” Submit pull request on GitHub Peer review & feedback Approve & merge changes Published to opentelemetry.io Community & Review Interaction 10 Everyone Contributor Approvers Everyone Maintainers Create Github issues Submits PR for review Review and support contributors Collaborative revisions, feedback on both directions Merge PR © 2026 Cloud Native Computing Foundation
11. Where the localization community lives GitHub CNCF Slack Bi-weekly Meetings • Issues & PRs • Questions & coordination • Discuss priorities • Code reviews • Meet community members • Unblock contributors • Localization work • New contributor intros • Coordinate concrete efforts You don't need to know everything before you join 11 © 2026 Cloud Native Computing Foundation
12. This works, but, the English documentation never stops changing 12 © 2026 Cloud Native Computing Foundation
13. What is toil? 13 © 2026 Cloud Native Computing Foundation
14. Localization is Never Finished English docs evolve The Moving Target • Pending translations • Content drift • New content • New languages added Automating Work vs. Value of "Toil" We can automate much of this workload, but is it all toil? • Minor Edits: Fixing a typo may teach little about OpenTelemetry concepts... • Onboarding Value: ...yet for new OSS contributors, simple tasks teach the contribution process without overwhelming them. © 2026 Cloud Native Computing Foundation
15. OpenTelemetry docs already automate some of the localization “toil” WHEN ENGLISH SOURCE CHANGES OCCUR: → Detect translation drift Flags out-of-sync localized files automatically whenever upstream docs change. → Update localization metadata automatically Maintains header metadata and sync tags without manual maintenance. KEY INSIGHT This is clearly toil . It does not create a meaningful learning opportunity. 15 © 2026 Cloud Native Computing Foundation
16. Where is the boundary between toil and getting started contributions? AI can easily automate all (if not all) good first issues spell find_in_page rocket_ sync_alt g_ Fix typos Detect drift Grammar, spelling & formatting fixes Flagging out-of-sync docs Manual today ✕ Automated today ✓ ← Less project knowledge 16 © 2026 Cloud Native Computing Foundation Bootstrap localization Update drifted page Translate new page Scaffolding new language setups Re-aligning modified content Initial full page translations Automated today ✓ Manual today ✕ Manual today ✕ More project knowledge →
17. So… what is toil? 17   01 / FIRST CONTRIBUTION 02 / LEARNING PROCESS 03 / AI CAPABILITY Is a repetitive task still toil if it is someone's first PR? Is translation toil if it helps someone learn the project? If AI can do the whole task, does that mean it should? © 2026 Cloud Native Computing Foundation 
18. AI-assisted localization 18 © 2026 Cloud Native Computing Foundation
19. What are other open source projects doing? 01 / AUTOMATED WORKFLOW  dlt AI workflow for docs Automated documentation processing integrated into pull request workflows. 19 © 2026 Cloud Native Computing Foundation
20. What are other open source projects doing? 01 / AUTOMATED WORKFLOW  dlt AI workflow for docs Automated documentation processing integrated into pull request workflows. 02 / HUMAN OVERSIGHT  Humans in the loop Mostly driven by agents, while human maintainers retain final review authority. 20 © 2026 Cloud Native Computing Foundation
21. What are other open source projects doing? 01 / AUTOMATED WORKFLOW  dlt AI workflow for docs Automated documentation processing integrated into pull request workflows. 02 / HUMAN OVERSIGHT  Humans in the loop Mostly driven by agents, while human maintainers retain final review authority. 03 / OPEN QUESTION  Where does a new contributor fit in the docs ? 21 © 2026 Cloud Native Computing Foundation
22. Experimenting with more automation in OTel Spanish localization 01 / AGENT CAPABILITY  Teaching agents localization Can we teach an agent how localization works? Yes! 02 / ENCODED KNOWLEDGE  Project conventions & rules Encode project conventions, terminology, workflow, and validations. 03 / KEY TAKEAWAY  Skills as another form of project documentation . 22 © 2026 Cloud Native Computing Foundation
23. Demo https://github.com/immavalls/otel-localization-skills 23 23 © 2026 Cloud Native Computing Foundation © 2026 Cloud Native Computing Foundation
24. But… who, if anyone, is learning?   02 / CONTRIBUTORS Automating beginner tasks Risk of not learning Maintainers can automate away beginner tasks. But, contributors can also delegate the task to AI, risking skipping the learning process.  24 01 / MAINTAINERS 03 / POLICY EXPECTATION OpenTelemetry docs AI contribution policy expects contributors to understand, review and own what they submit. © 2026 Cloud Native Computing Foundation
25. Automate the toil, not the community 25 © 2026 Cloud Native Computing Foundation
26. Automate the Toil, not the Community    01 / WHAT SHOULD WE AUTOMATE WITH AI 02 / WHAT SHOULD CONTRIBUTORS LEARN • What is genuinely toil? • OpenTelemetry concepts and terminology. 03 / WHAT CAN REPLACE LOCALIZATION AS ENTRY POINT • What improves accessibility and project health if automated? • What are we only keeping manual because it used to be a low-friction entry point? 26 © 2026 Cloud Native Computing Foundation • How the contribution workflow works. • How to question and validate AI output. • How to interact with project members. Localization is only one example. What other “good first contributions” are we about to automate away? • Learn the project, how to contribute, build trust and grow.
27. Automate the Toil, not the Community 27    04 / MY ENTRY POINT 05 / THE TRADEOFF 06 / OPEN QUESTION Localization was my own entry point into OpenTelemetry, so I’m obviously biased by that experience. I can see the benefit of automating more of it. I don’t know yet where the right boundary is... I can also see what we might lose in the contributor journey. We risk not just automating the friction, but the contribution itself. ...and I’d really like to hear what you think. © 2026 Cloud Native Computing Foundation
28. Q&A https://eyeveebee.dev/automating-the-toil-not-the-community-ai-assisted-opentelemetry-docs-localization
29. Thanks