GTM AI工程师
GTM AI Engineer
关于 Customer.io
超过 9000 家公司——从初创企业到全球品牌——使用我们的平台每天发送数十亿封电子邮件、推送通知、应用内消息和短信。Customer.io 驱动自动化沟通，人们真正希望收到的沟通。我们帮助团队利用实时行为数据发送更智能、更相关的信息。
关于该职位
大家好，我是 Nick Guardino，Customer.io 的市场运营高级总监，我正在寻找一名 GTM AI 工程师，帮助我们将 AI 代理从初步原型提升为 GTM 团队可以依赖的生产级系统。
在过去的 3.5 个月中，我们的 GTM AI 团队已经构建了 11 个代理，现在已在生产环境中运行，服务 120 名活跃用户，每天大约进行 25,000 次工具调用——全部通过定制 CLI 实现，但这些 CLI 并未设计为可扩展。我们认为，当企业有效部署 AI 代理时，可以更高效地运作并走得更远，而我们正在实时探索这一过程。这个职位负责扩展和可靠性层，让团队其他成员能够继续构建。
我们重视的品质
- 工匠精神胜过短期收益——热衷于打造稳健的系统，而不仅仅是让系统运行。
- 有自己观点——希望你问“为什么这样构建？”并能提出具体的改进建议。
- 能接受犯错——你快速尝试新事物，并能同样快速地调整方向。
- 默认 AI 原生——你更倾向于让 Claude 编写和测试代码，而不是手动编写。
你的工作内容
- 从 Alpha 到生产环境的全权负责——将新构建的代理从 v1 版本提升至生产就绪状态：编写评估，强化提示和工具定义，并将原型结构重构为干净、高质量的生产代码。
- 可观测性与稳定性——在代理集群中实现追踪、日志记录和监控，以便在我们的 120 多名 GTM 用户之前发现回归问题。
- 部署与基础设施——建立并维护将代理交付给用户的管道，确保无意外发生，负责 GCP 架构中的环境一致性与发布规范。
- 数据库管理——负责代理如何安全且可扩展地读取和写入我们的数据库。
- 集群维护——监控生产环境中的代理行为，处理问题，并持续推出改进，使代理变得不可或缺而非仅仅有用。
- 与 GTM AI 团队紧密合作——让我们的代理开发者专注于他们的零到一领域，而你负责扩展和可靠性。
- 带来外部视角——告诉我们我们哪里做得不对
查看英文原文
About Customer.io
Over 9,000 companies — from scrappy startups to global brands — use our platform to send billions of emails, push notifications, in-app messages, and SMS every day. Customer.io powers automated communication that people actually want to receive. We help teams send smarter, more relevant messages using real-time behavioral data.
About the role
Hi, my name is Nick Guardino, Senior Director of Marketing Ops at Customer.io, and I'm looking for a GTM AI Engineer to help take our AI agents from scrappy prototypes to production-grade systems our GTM teams can depend on.
Over the last 3.5 months, our GTM AI team has built 11 agents now running in production, serving 120 active users and roughly 25,000 tool calls a day — all through custom CLIs that work, but weren't built to scale. We believe businesses can run leaner and go further when they deploy AI agents well, and we're figuring out that motion in real time. This role owns the scale and reliability layer so the rest of the team can keep building.
What we value
- Craft over quick wins — energized by making things robust, not just making them work.
- A point of view — we want someone who asks "why did you build it that way?" and has a specific answer for what to do instead.
- Comfortable being wrong — you try things fast, and you course-correct just as fast.
- AI-native by default — you'd rather direct Claude to write and QA code than write it by hand.
What you'll do
- Alpha-to-production ownership — take newly built agents from working v1 to production-ready: write evals, harden prompts and tool definitions, and refactor prototype scaffolding into clean, production-quality code.
- Observability & stability — instrument tracing, logging, and monitoring across the agent fleet so we catch regressions before our 120+ GTM users do.
- Deployment & infrastructure — stand up and maintain the pipelines that get agents into users' hands without surprises, owning environment parity and release hygiene on our GCP stack.
- Database administration — own how agents read from and write to our databases, safely and at scale.
- Fleet maintenance — monitor agent behavior in production, triage issues, and ship the steady stream of improvements that make an agent indispensable rather than just useful.
- Partner closely with the GTM AI team — free up our agent builders to stay in their zero-to-one wheelhouse while you own scale and reliability.
- Bring an outside point of view — tell us what we built wrong, and back it up with a plan to fix it.
What we're looking for
- 4+ years in software engineering, platform/infrastructure engineering, or a similarly technical role, ideally at a B2B SaaS company.
- Comfort working in the terminal daily, with a real GitHub PR-and-deploy workflow.
- Cloud hosting experience — Kubernetes, VMs, or equivalent; Google Cloud experience strongly preferred.
- Experience with how agents interact with and edit databases, including some hands-on database administration.
- Familiarity with LLM-based application development: prompt engineering, tool/function calling, evals, and modern agent frameworks.
- Ability to prompt AI tools correctly, review their output critically, and direct them to QA their own work — this is an AI-native role, not a hand-coding one.
- Scrappy enough to iterate fast, disciplined enough to think about consistency once something needs to scale.
- A point of view, and the willingness to ask "why did you build it that way?"
Compensation & Benefits
We believe in transparency. Starting salary for this role is $140,000–$170,000 USD (or equivalent in local currency) depending on experience and subject to market rate adjustment.
We know our people are what make us great, and we’re committed to taking great care of them. Our inclusive benefits package supports your well-being and growth, including 100% coverage of medical, dental, vision, mental health, and supplemental insurance premiums for you and your family. We also offer 16 weeks paid parental leave, unlimited PTO, stipends for remote work and wellness, a professional development budget, and more.
See full benefits here →
Our Process
No gotchas, no trick questions - just a clear, human process designed to help both of us make an informed decision.
- 30-minute Zoom call with Recruiter
- 45-minute Zoom call with Hiring Manager
- 45-minute Zoom call with a member of the GTM AI team
- 60-minute Live technical session with the team
All final candidates will be asked to complete a background check and employment verifications as part of our pre-employment process.
Customer.io recognizes the stifling impact of systemic injustice on diverse communities. We commit to using our influence to increase inclusion and equity within the tech industry. We strive to build an inclusive team culture, implement bias-free hiring practices, and develop community partnerships to expand our global impact.
Zoom is the only video conference platform that we use, virtual interviews will be conducted using the video capability (i.e., not via the chat), and offers will be extended in writing on official Customer.io letterhead. Please be vigilant in all of your job search activity, and if you have any questions please contact jobs@customer.io.
Join us!
Check out our careers page for more information about why you should come work with us! We believe in empathy, transparency, responsibility, and, yes, a little awkwardness. If you’re excited by what you read — apply now.