XConf North America 2021
who it's for
software architects, platform engineers, data engineers, machine learning engineers, devops engineers, engineering managers, product managers
topics
evolutionary architecture and fitness functions; continuous delivery and CI safety nets; consumer-driven contracts for services; Azure Kubernetes Service deployments; Grafana dashboards and observability stacks; CD4ML model pipelines and versioning; data mesh testing and data product packaging; federated computational governance; architecture decision records (ADRs); explainable AI methods and counterfactuals; developer portals and self-serve platforms; value stream optimization and DORA signals
| title | author | length |
|---|---|---|
| The past, present and future of Architecture | Rebecca Parsons | 58m |
Argues that architecture must be evolutionary — explicitly guided and measurable across reliability, scalability, security and maintainability — enabled by… | ||
| Observability in digital transformation | Ajay Chankramath, Sean Alvarez | 46m |
A mid‑market finance rollout builds a low-cost observable aggregator that preserves source systems while offering a single pane of glass, ingesting event… | ||
| Getting Machine Learning applications into production using MLOps and CD4ML | Eric Nagler | 14m |
87% of data science projects never reach production, so CD4ML lays out a concrete six-step pipeline — model building, evaluation, productionization, testing… | ||
| Product innovation, not derivation | Dave Ho | 19m |
Argues that limited technical acumen forces product teams into copycat features and lower success rates—about 70% failure for new features/products—and gives… | ||
| Lightweight architecture governance in the large | Imran Khan, Leo Mattos | 16m |
Describes a lightweight governance framework proven on a multi‑year enterprise modernization that moves teams from 5 to 15+ and across acquisitions, showing… | ||
| Using Explainable AI to garner trust and overcome adoption inhibitors | Ashley Avery | 26m |
95% cite cultural, organizational and process barriers to AI adoption, and trust—not just transparency—is the bottleneck. Presents concrete XAI methods: shop… | ||
| Augmented Reality Basics | Paris Kaman | 10m |
Augmented reality depends on computer vision not hardware hype and uses three concrete anchoring methods: image targets, face meshes, and anchors from feature… | ||
| Data Mesh testing: An opinionated view of what good looks like | Sofia Tania | 20m |
A practical 15-minute roadmap that prescribes specific tests per product type: unit and component tests for transformation logic, data quality checks and… | ||
| Optimizing the engineering value stream | Prem C | 15m |
Argues that elite teams converge on three traits: rapid, reliable, repeatable delivery and that the State of DevOps four metrics — deployment frequency, lead… | ||
| Data Mesh: Lessons from the trenches | Sina Jahan, Storm Heidinger | 44m |
Large Fortune‑100 healthcare payer tries to move analytics to the cloud with a goal of ~1,000 assets migrated, but finds success only by building a custom… | ||
| The uncanny valley of hybrid working | Tom Marsh | 14m |
Argues that team-level dynamics determine whether co-location, remote, or hybrid works and that hybrid often loses most benefits: ideal co-location scales to… | ||
| I don't think these metrics mean what you think they mean! | Ranbir Chawla | 14m |
Argues that DORA metrics are lagging indicators and instead recommends leading signals: measure cycle time and inner cycles, limit WIP, track error budgets and… | ||
| Developer portal | Tim Cochran, Ama Asare | 15m |
Joiners often waste hours hunting docs across Confluence, Slack and drives; a developer portal claims to stop that by centralizing a Backstage service catalog… | ||
| Scaling a crypto currency exchange startup | Rickey Zachary | 47m |
Explains that crypto exchanges must handle 3,000% traffic spikes and roughly 300× user growth from single social posts, so availability, consistency and… | ||
| Thoughtworks culture: 10,000 and beyond | Venki Ponniah, Patricia Mandarino | 36m |
Thoughtworks claims a 28‑year culture that produced concrete tools and practices: Selenium and CruiseControl and a microservices paper that become industry… | ||
| nothing matches | ||