Technical Writer, Open Source

Seeking a Technical Writer, Open Source to join the Core ML Platform team, which owns developer experience and pipeline authoring, the Ray-native execution engine, and the platform's open source strategy. Responsibilities Own the documentation surface for new feature releases Track PRs and changes as the platform ships frequently, working directly from PR descriptions, code, and engineering discussions; and refresh the corresponding user guides, API references, and changelogs before or immediately after merge. Monitor open PRs and issues for documentation-affecting changes across the Uniflow pipeline authoring layer, Ray execution engine, and control plane. Write release notes and CHANGELOG entries that translate engineering changes into clear, user-facing impact statements. Coordinate with TLs and engineers to ensure docs are in-sync with each release milestone. Produce community-facing content (user guides, contributor guides, blog posts) Create the content that helps internal developer community and external open-source adopters succeed. Write and maintain the Sandbox Setup guide, Getting Started with ML Pipelines, and end-to-end tutorials using API and Uniflow YAML DSL. Expand and improve the Contributor Guide, issue triage norms, and development environment setup. Author promotional blog posts and partner-facing content that position the platform within the broader MLOps ecosystem (Ray, Kubernetes-native ML, distributed training on OKE). Respond to community documentation feedback surfaced in GitHub Discussions and Issues 3. Enforce consistency across all documentation Serve as the documentation quality owner. Review all doc-labeled pull requests for style, accuracy, and consistency before merge. Establish and maintain a style guide covering terminology (e.g., Uniflow, pipelinerun, triggerrun, CRD, RayJob, sandbox), code formatting, and conceptual framing. Audit the /docs directory and website periodically for stale content, broken cross-references, and gaps between documented behavior and current code. Define the canonical source of truth for each major concept, from pipeline authoring to model serving, and enforce single-source practices across guides, API references, and READMEs. 4. Build an agent-assisted internal documentation system Architect a documentation agent harness, a Uniflow pipeline that watches merged PRs, extracts feature context from code and PR descriptions, and drafts or flags documentation updates for human review. Define structure source of truth schemas (e.g., for feature capabilities, API surfaces, config parameters) that agents can read from and write to reliably, reducing hallucination and drift. Collaborate with team members to integrate the doc-agent into the existing CI/CD and release workflows. Evaluate and adopt appropriate tooling, Claude Code, Inspect AI, custom Uniflow tasks to automate coverage gap detection, link validation, and changelog generation. Write documentation about the documentation system itself: how it works, how to extend it, what requires human review. Basic qualifications 3+ years of technical writing experience, with at least 2 years on an open source or developer-facing platform project. Proven track record authoring user guides, contributor guides, getting-started tutorials, and API references for software used by engineers not just non-technical users. Experience keeping documentation in sync with a fast-moving codebase: reading diffs and PRs to identify documentation impact, not waiting to be told. Strong writing craft: the ability to explain complex distributed systems concepts (orchestration, pipeline DAGs, CRDs, training loops) clearly and concisely without sacrificing accuracy. Comfortable working in GitHub: PR reviews, issue triage, branch management, Markdown. Preferred qualifications Familiarity with ML infrastructure concepts: model training (PyTorch, Ray Train, PyTorch Lightning), pipeline orchestration (Argo, Flyte, Kubeflow, or similar), model serving, feature stores. Experience writing promotional engineering blog posts that attract developer audiences, ideally for an OSS project you were embedded in. Some hands-on exposure to coding agents, LLM tooling, or automated documentation generation (e.g., building or evaluating doc-generation pipelines with Claude, GPT, or similar). Prior experience at a platform or infrastructure company (e.g., cloud provider, MLOps vendor, data tooling company, or large tech platform team). Familiarity with Kubernetes and Helm at the conceptual level, enough to accurately document sandbox setup, CRD behavior, and cluster configuration steps. The estimated pay range for this position is USD $45.00/Hr - USD $50.00/Hr. Exact compensation and offers of employment are dependent on job-related knowledge, skills, experience, licenses or certifications, and location. We also offer comprehensive benefits. The Talent Acquisition Partner can share more details about compensation or benefits for the role during the interview process. Milestone Technologies