平台工程师
Platform Engineer
平台工程师（中级+）
Claid.AI 正在将生成式 AI 带入企业级应用。平台团队负责云基础设施和围绕安全交付这些管道的开发者体验。小团队，大范围的工作内容，从第一天起就拥有真正的主导权。
你将参与的工作
在 GCP、AWS 和 GPU/neo-cloud 上使用 Terraform 运行我们的基础设施即代码。在多仓库设置中保持 GitHub Actions 的 CI/CD 健康。解决任何阻碍产品和机器学习工程师的问题——本地开发、预览环境、密钥、部署、可观测性。参与轮班值班，负责运行手册，并跟进事件直至解决。
你将与我们的平台/工程负责人密切合作——有 16 年经验，之前在 Grammarly 和 Amazon Ring 管理过平台。他一直在独自承担大部分工作；你是这个两人职能中的第二位成员，从第一周起就拥有真正的主导权，而不是作为初级人员跟随他人处理问题。
技术栈
Terraform（日常大量使用 —— 模块、状态、漂移），Kubernetes，GitHub Actions，GCP + AWS，GPU/neo-cloud。
- 同时涉及：Cloudflare，RabbitMQ，SQL 和 NoSQL 数据库，Grafana + Mimir 监控。
- 加分项：Argo / GitOps。
我们寻找的人选
- 中级及以上，2–4 年真实平台/DevOps/SRE 工作经验。
- 你更倾向于从头到尾负责并亲自交付，而不仅仅是边缘贡献。
- 你不需要了解技术栈中的每一项，但你渴望学习。
- 你曾深夜调试被阻塞的部署并成功解决（或类似经历）。
我们重视的方面：
- 你曾在 GCP 或 AWS 上运行过工作负载，能够独立思考 IAM、网络和成本问题。
- 你熟悉 SQL/NoSQL 数据库。
- 你了解 Terraform —— 模块、状态和漂移。
- 你曾运营过生产环境的 CI/CD 流水线 —— 构建阶段，修复不稳定的部分，调试失败的部署。
- 你用 Python、Go 或 Rust 编写工具/脚本。我们大量使用 AI，但你可以不依赖 AI 编写代码 —— 我们会验证。
- 你自动化重复劳动，而不是手动一次性操作，而是交付小而可逆的变更。
初创公司风格：适应模糊性，主动承担未被关注的问题，把事情记录下来，让下一个人不用重新思考。
值班安排
是的，与平台负责人共享 —— 具体轮班还在确定中。团队中的其他工程师也担任第一响应者，因此你不是唯一的防线。
我们提供的福利
- 美国母公司股票期权 —— 标准配置
查看英文原文
Platform Engineer (Mid+)
Claid.AI is delivering GenAI to enterprise. The platform team owns the cloud footprint and the developer experience around shipping those pipelines safely. Small team, lots of surface area, real ownership from day one.
What you'll work on
Run our infrastructure as code in Terraform across GCP, AWS, and GPU/neo-clouds. Keep CI/CD healthy on GitHub Actions across a multi-repo setup. Fix whatever is slowing product and ML engineers down — local dev, preview environments, secrets, deploys, observability. Take part in on-call, own the runbooks, and follow incidents through to a fix.
You'll work closely with our Head of Platform/Engineering — 16 years in, previously ran platform at Grammarly and at Amazon Ring. He's been carrying most of this solo for a while; you're the second pair of hands on a two-person function, with real ownership from week one rather than a junior seat shadowing someone else's calls.
Stack
Terraform (heavy daily use — modules, state, drift), Kubernetes, GitHub Actions, GCP + AWS, GPU/neo-clouds.
- Also in the mix: Cloudflare, RabbitMQ, SQL and NoSQL datastores, Grafana + Mimir for monitoring.
- Nice to have: Argo / GitOps.
Who we're looking for
- Mid+ 2–4 years of real platform/DevOps/SRE work.
- You prefer to own things end-to-end and ship them yourself, not just contribute around the edges.
- You don't need to know every item in the stack, but you are hungry for knowledge.
- You've debugged a blocked deploy at 11pm and come out the other side (or similar).
What we care about:
- You've run workloads in GCP or AWS and can reason about IAM, networking, and cost without hand-holding.
- You know your way across SQL/NoSQL databases.
- You know Terraform — modules, state, and drift.
- You've operated a CI/CD pipeline in production — built the stages, fixed the flaky ones, and debugged the failed deploy.
- You write tools/scripts in Python, Go, or Rust. We use AI heavily, but you can write code without it — and we'll check.
- You automate the toil and ship small, reversible changes instead of manual one-offs.
Startup-shaped: comfortable with ambiguity, picks up unowned problems, writes things down so the next person doesn't have to re-figure them out.
On-call
Yes, shared with the Head of Platform — exact rotation is still being finalized. Engineers across the team also serve as first responders, so you're not the only line of defense.
What we offer
- Stock options in the US parent company — standard terms, comparable to what you'd see at a Silicon Valley startup.
- A genuinely stable team for a 9-year-old company — outside of a few very recent hires, the team has stayed an average of 2+ years.
- A flexible, output-based culture — roughly 21 vacation days a year, not rigidly tracked.
- Direct mentorship from a Head of Platform well known in the (Ukrainian-speaking) DevOps community.
- Founder access: every single hire at Claid, including this one, gets a 30-minute interview slot with our CEO, Vlad — we're hands-on, not a black box.
Location
LatAm-based, remote, preferred. Working hours need to overlap meaningfully with US business hours (Pacific Time). Open to candidates based in Europe.
Originally posted on Himalayas