Show Notes
Seeing GraphQL everywhere at SEEK and wondering what it actually buys you? In this episode of SEEK Bytes, host Seamus Kearney is joined by Nik Skoufis, Reyna Tan and special guest Mark Cheesman to unpack how SEEK uses GraphQL across candidate and hirer systems – from the early “shared free‑for‑all” graph to today’s curated, high‑scale platform. They dig into what GraphQL really is (a query language and graph executor, not “just an API”), how resolvers and data sources work under the hood, and why SEEK leans on GraphQL to power experience APIs for web and mobile instead of hand‑rolling bespoke REST layers.
This episode's special guest: Mark Cheeseman (SEEK Staff Engineer)
In this episode, we explore: • What GraphQL actually is (beyond “an API”) – a query language and execution engine that turns your request into a graph and traverses it, not just “REST with curly braces” – and why it doesn’t even need HTTP under the hood. • Why SEEK moved from REST experience APIs to GraphQL – unifying messy mobile and web backends, replacing one-off “backend-for-frontend” services with a flexible query layer so clients can build their own experience APIs on every request. • Resolvers, data sources & performance at scale – how resolvers map to REST APIs, S3, caches and in-memory data, why SEEK prefers field-level resolvers and data-source caching, and how graph theory and query-complexity limits protect against “select * from everything” type queries. • Security, rate limiting & degraded behaviour – preventing scraping and abusive queries, using query-level rules on top of IP-based rate limiting, and designing for graceful degradation when upstream services are slow or down instead of treating the site as simply “up or down”. • Custodianship, observability & culture – why a shared graph needs an owning team, how SEEK uses OpenTelemetry, Datadog and custom metrics to see who’s calling what, find hotspots and retire unused fields, and how pairing and “graph custodians” help other teams onboard safely.
If you’re a frontend, backend or full‑stack engineer trying to decide when GraphQL is worth it – or just want to understand SEEK’s graph stack better – this episode is for you.
🔔 Follow the SEEK Bytes podcast so you never miss a new episode
More from SEEK Bytes
This special Season 3 finale of SEEK Bytes – a podcast for engineers by engineers – dives into one of the biggest AI engineering shifts of the moment: spec-driven development with GitHub Spec Kit. If you’ve been wondering how teams are using AI for more than just code completion, this episode unpacks a workflow designed to help engineers plan better, reduce drift and build higher-quality software faster.
What happens when you give 100 high school students four days, real SEEK mentors and a crash course in how tech actually works? In this episode of SEEK Bytes – we're joined by special guests Asha Ioculari and Leon Belobrov and go behind the scenes of Camp SEEK, SEEK’s immersive program designed to open the doors of tech to girls and non-binary students and show them there’s far more to this industry than just writing code. From design thinking and user research to pitching in front of 200+ people and senior SEEK leaders, this is a brilliant look at how SEEK is building confidence, broadening horizons and helping shape a more inclusive future for tech.
Ever shipped a UI that “works” but still feels clunky, janky or confusing to users? In this episode of SEEK Bytes, Elliott and Will chat with lead engineer Eddy Nguyen about Bleeds – a simple but powerful way to design every UI state your users actually experience, not just the happy path.
AI might look simple on the surface — just send a prompt, get a response — but behind the scenes, there’s a whole platform challenge to solve. In this episode of SEEK Bytes – we’re joined by Thet Ko, Senior Staff Engineer in SEEK’s AI Platform Services team, to unpack how SEEK is building an AI Gateway to centralise AI calls, improve governance and make it easier for teams to build with AI without reinventing the foundations every time. From cost visibility and PII protection to guardrails, resilience and shared memory across agents, this episode lifts the lid on the platform work that makes enterprise AI actually usable at scale.
AI is moving fast — but are we building it responsibly? In this episode of SEEK Bytes – we sit down with Sreyna Rath, CEO and co-founder of jaimee.ai, who arrived in Australia as a refugee and is now running her own AI company with the aim to create something truly transformative for women. Elliott sits down with Reyna to unpack her journey and the real-world risks of AI bias, why safety can’t be bolted on at the end, and what developers can do to build tools that are fairer, safer and more human from day one. From biased hiring systems to sexualised image outputs and AI companions designed for vulnerable users, this conversation is a powerful reminder that AI doesn’t just reflect technology — it reflects us.