Articles

Explore our collection of technical articles, tutorials, and interactive learning experiences

Abstract diagram of partitioned resource pools on a shared cluster, representing multi-tenant isolation boundaries

Multi-tenancy on a shared AI platform — quotas, fairness, and the noisy-neighbour problem

How to share a Kubernetes-based AI platform across teams without letting one tenant starve the others — covering isolation models, quota mechanics, priority classes, and network boundaries.

June 9, 20269 min read

Six interconnected role cards on an AI platform team, showing Platform Engineer, MLOps Engineer, ML Engineer, Data Scientist, DevOps/SRE, and Data Engineer

The six roles on an AI platform — what each does and what each is fluent in

A practical breakdown of the six roles on an AI platform team: job-to-be-done, skill bar, day-one tools, and the anti-patterns that surface when a role is mis-hired.

Jun 9, 2026AI Platform Engineering & MLOps

Layered diagram showing an AI Platform team's three ownership layers: substrate, ML primitives, and application support

What an AI Platform Team Actually Owns — and What It Does Not

A precise definition of the AI Platform team as a platform-as-a-product function, its three ownership layers, its accountability boundary, and how it hands off to the MLOps engineer.

Articles

Multi-tenancy on a shared AI platform — quotas, fairness, and the noisy-neighbour problem

The six roles on an AI platform — what each does and what each is fluent in

What an AI Platform Team Actually Owns — and What It Does Not

More Articles

Four organisational patterns for shipping ML — and when each one breaks

Governance and lineage — answering the four questions a regulator will ask

SBOM and Signing for ML — Extending Supply-Chain Security from Container Images to Model Artifacts

When ML breaks: an incident-response playbook for production models

Supply-chain security for ML — SLSA, signing, and admission verification, end to end

Dynamic Resource Allocation: What Changes When Devices Become First-Class

FinOps for AI: The Showback-to-Chargeback Ladder and Unit Economics That Actually Work

Training workloads on Kubernetes — operators, gang scheduling, and checkpointing

Eval as a Test Suite: LLM-as-Judge in CI Without Flaky Merges

Responsible AI and Governance — Turning the EU AI Act into a Platform Checklist

HAMi — fractional GPU on the GPUs you actually have

Why share a GPU? The economics, the mechanics, the four mechanisms

Pure-cloud, on-prem, hybrid, air-gapped — the deployment-context spectrum

AI Platform maturity — five levels and the single move that unlocks each

Serving patterns for production ML — runtimes, routing, and the autoscaling signals that matter

Where this goes next — the open problems on a 2026 AI platform

The AI gateway — what it is, when you need one, and where it sits