What is the Role

You are the person a customer calls when their Elasticsearch platform is the problem, or when they need one designed properly the first time.

That means you decide the shape of a cluster and then defend that decision — node topology, shard strategy, data tiers, lifecycle and retention — with the sizing arithmetic to back it up. It also means you can follow a performance problem beyond Elastic itself. We need someone whose instinct is to look down the stack and sideways to whatever else shares the infrastructure, rather than stopping at the cluster API.

What You Will Be Doing

• Design and build Elasticsearch platforms for customers across self-managed, orchestrated and managed deployments, including air-gapped and regulated environments
• Own sizing and capacity planning as a derivation from workload characteristics — ingest rate, document shape, retention, query concurrency, growth — validated by benchmark rather than estimated from a form
• Run health checks and remediation on production estates, and redesign sharding, mapping and index strategies where the original design no longer holds
• Lead major-version upgrades, snapshot and disaster recovery strategy, and multi-cluster architectures
• Diagnose performance problems across the full stack — Elastic, the JVM, the operating system, storage and network — and across adjacent systems in the data path
• Work directly with customer architects, platform owners and engineering leadership; run discovery, present designs, and own the technical outcome of an engagement end to end
• Build reusable tooling, automation and IP that makes the next engagement faster, including AI-assisted tooling for diagnosis, configuration and reporting
• Mentor other consultants and contribute to the GCC's technical practice

What You Bring

Elastic depth — non-negotiable

We are not looking for someone who will learn Elastic on the job.

• 8+ years hands-on with the Elastic Stack, substantially on cluster and platform work rather than dashboarding or pipeline authoring
• Shard strategy you can derive, not recite — how you arrived at a shard count and size for a given workload, what you traded off, and what you got wrong the first time
• Index and mapping design, and why those choices drive cluster performance
• Lifecycle and retention across data tiers, including the cost and latency implications of each
• Snapshot, restore and disaster recovery designed and executed, not just configured
• Major-version upgrades in production — the sequencing, the breaking changes, the rollback position
• More than one deployment model: self-managed plus at least one managed or orchestrated option
• Cluster stability and performance troubleshooting — heap and garbage collection, circuit breakers, thread pool rejections, shard allocation, expensive queries
• Elastic Certified Engineer, or the demonstrable knowledge to obtain it within your first quarter
• Strong Linux and operating system fundamentals, with solid grounding in storage and networking
• A working understanding of how infrastructure conditions translate into Elastic symptoms, and how infrastructure choices constrain what a deployment can do
• Experience troubleshooting on shared infrastructure, where the cause of your problem is another workload's behaviour and resolving it means engaging the team that owns it

#LI-DS1