Senior Engineering Manager - Platform Engineering
Leads Zappi's five-person Platform Engineering team in Cape Town, running core AWS infrastructure and AI workloads (self-managed Kubernetes via kOps, Terraform, GitOps) and building the in-house developer platform. People-management role with hands-on coding, incident response, and shaping how AI agents work safely on the platform.
- Compensation
- ZAR 1.1M–ZAR 1.3M / yr
- Location
- Cape Town , South Africa
- Work setup
- Hybrid
- Seniority
- Senior
- Employment
- Not specified
- Experience
- Not specified
- Category
- Devops
- Posted
Skills
Must have
- Strong platform skills: Kubernetes, cloud infrastructure (ideally AWS), infrastructure as code, CI/CD and observability
- Comfortable debating architecture with senior engineers
- Excited and curious about AI in engineering; personally uses tools like Claude Code or Codex and has thought about AI working safely on real systems
- High standards for running production: availability, upgrades, security, cost and capacity
- Calm in an incident and runs on-call in a way people don't dread
- Product sense: treats engineers as customers and knows a tool isn't done until people use it
- Genuine enjoyment of managing people: growing engineers, giving honest feedback, helping a team through change
- Around ten years in software or infrastructure engineering
- At least three years managing platform, SRE or infrastructure engineers
Nice to have
- Experience running self-managed Kubernetes (ideally with kOps)
- Experience with GPU or AI infrastructure
Full description
Zappi helps some of the world’s biggest brands test ideas before they spend money bringing them to life. That means new products, ads, names, packaging and flavours. Our customers make big decisions on our data, so it has to be accurate, fast and affordable.
We make around $100M in revenue. Our product engineering team is about 90 people, based in London and Cape Town.
Why this role exists
AI has changed how we write code. Throughput at Zappi keeps going up, and the bottleneck is moving to everything around the code. That means environments, config, deploys and access, plus keeping more and more of it running safely and cheaply.
That’s the space Platform Engineering works in. The team runs our core infrastructure and AI workloads, and builds the developer platform our engineers use to set up, deploy, configure and monitor their apps. Infrastructure and app setup are defined as data, which is what makes the platform flexible. In the last month, teams and automations completed over 20,000 deployments, 5,600 of them to production, all without needing us. Anything we make easier gives time back to the rest of engineering.
Next, we want agents working on that same platform. Picture one that spots an error in production, spins up its own isolated environment with the right context, reproduces the issue and works out what’s wrong. It then either hands over a clear diagnosis or opens a pull request with the fix, and we can see and review every step. We can only do that if the platform is built for it.
Our Platform Engineering team runs that platform and builds what comes next. We’re looking for someone to lead it.
You’ll lead a team of five platform and SRE engineers, working as one team across two areas.
Core Infrastructure.** Keep our infrastructure and AI workloads reliable, secure, current and cost-effective. That means self-managed clusters built with kOps on EC2, everything in Terraform, and GitOps deploys. We already run GPU workloads and self-hosted models in production, and you’ll help teams work out when open-source models beat calling an API.
Developer Experience. Build the tooling that lets teams ship without us in the loop, for people and agents alike. Our in-house developer platform sits at the centre of it. Several projects are already under way:
- Getting environments the data they need, so nobody loses time waiting for a place to build & test.
- Moving app setup and configuration out of code and into data, so teams can change it without a pull request.
- Making a new environment something you ask for and get in about ten minutes.
- Making sure anything an engineer can do to their app, an agent can do too, with the correct safety guardrails.