At a glanceSummarised by Seekless from the posting.
Must have6
Set the direction for a substantial piece of engineering work, delivered it and got other teams to adopt it
Built production instrumentation and telemetry integrations yourself
Defined both product-level and technical-level SLAs, SLOs and SLIs
Used SRE measurements such as error budgets, burn rates and MTTA/MTTR/MTTM
Built self-managed observability systems, such as OTEL, Prometheus, Grafana, Loki, Mimir, Alloy, Elasticsearch or Kibana
Worked with Kubernetes at scale, enterprise multi-tenancy, inference or asynchronous agent workflows
Eligibility1
Experience in AI is not required but mentioned as not needed
Skills
OTEL
Prometheus
Grafana
Loki
Mimir
Alloy
Elasticsearch
Kibana
Kubernetes
About the role
You’ll own observability across H: the strategy, technical design and rollout. Your work will give visibility of the entire stack and surface the information the company relies on to make key decisions and quantify progress.
About the company
H builds computer-use agents that take actions in browsers and desktop applications. This role covers every technical system we run.
What you own
The whole problem space. This is a lead role that stays hands-on: you will set the direction, design the stack, lead others through building it and work with internal teams on adoption. You’ll be responsible for ensuring our teams measure what matters, and for learning our products and services in depth.
What you’d be doing
•
Setting one observability strategy that can be applied across teams and technical stacks
•
Owning the implementation of the systems that manage metrics, logs, traces, alerts and dashboards
•
Defining SLAs, SLOs and SLIs with our product, research and infrastructure teams
•
Building end-to-end visibility of system dependencies, providing visualisations for users and context for agents
•
Automating how we track coverage, reliability, alert quality and the cost of execution
The first year
By month 3, every H product has SLAs, SLOs and SLIs defined and measured in code, with compliance reports produced automatically. By month 6, a single complete solution covers metric collection, log processing and tracing. By month 12, every system at H uses the same SDKs and interfaces, whether for alarms, metrics, logs or traces.
Who you’d work with
You’ll work with every team at H, across infrastructure, product, engineering and research. Our VP of Engineering is the hiring manager.
What we think it takes
Likely a good fit if you
•
Have set the direction for a substantial piece of engineering work, delivered it and got other teams to adopt it.
•
Have built production instrumentation and telemetry integrations yourself.
•
Have defined both product-level and technical-level SLAs, SLOs and SLIs.
•
Have used SRE measurements such as error budgets, burn rates and MTTA/MTTR/MTTM.
•
Have built self-managed observability systems, such as OTEL, Prometheus, Grafana, Loki, Mimir, Alloy, Elasticsearch or Kibana.
•
Have worked with Kubernetes at scale, enterprise multi-tenancy, inference or asynchronous agent workflows.
You do not need any experience in AI. If you match most of this but not every point, don’t hesitate to apply.
How we hire
Four stages: a 30-minute call with our Talent team, a 60-minute technical assessment, a 60-minute system design interview and a closing 30-minute conversation with the hiring manager.
Our process is tailored to the requirements for this specific role: the system design interview and the technical assessment closely match the work you would be delivering from day one.
Practicalities
The role is hybrid, with three days a week in either our Paris or our London office and occasional travel between the two. We offer a competitive package.