Notes on data, AI, IT
and security
No marketing fog. The way I think about real problems with founders and managers.
GDPR data inventory is not a legal task
Why the personal data register that GDPR requires is operationally useful - and how to build it properly before the regulation comes into force.
AI readiness: what companies confuse with actual preparation
Why the gap between interest in AI and operational readiness to deploy it is much larger than it appears after a conference or a demo.
The real cost of moving to microservices
What gets overlooked when companies plan a migration from a monolith to a microservices architecture, and how to assess those costs before the work starts.
A data pipeline is a production system, not a script
Why companies lose trust in their analytics when they treat data pipelines as one-off tasks rather than operated systems.
Meltdown and Spectre: when the CPU layer became a security problem
What processor vulnerabilities mean for executives, and why they change the conversation about security at the infrastructure level.
Bitcoin, blockchain, and what business actually needs from either
While bitcoin sets records, I separate two different conversations: cryptocurrency as a speculative asset and blockchain as a business tool.
Technical debt: how to talk about it with non-technical leadership
Why technical debt is not just a technical problem, and how to discuss it with boards and owners in a way that leads to decisions rather than defensiveness.
GDPR takes effect in six months: what companies with EU exposure need to do
The European data protection regulation goes live in May 2018. Companies with European customers or offices are required to comply, regardless of where they are based.
Kubernetes: what the hype omits about the operational side
Kubernetes solves real problems at scale. It also introduces a new operational surface that most teams are not ready for. A realistic look before you commit.
API-first is a business decision, not a technical one
Why the API-first approach is about business architecture rather than development practice, and how it affects company agility over three to five years.
Data quality: four metrics that are worth tracking in practice
Most data quality programs stall because the metrics are too abstract. Here are four concrete measurements that show up problems early and connect to business outcomes.
Why ML teams keep rebuilding the same data pipelines
The hidden cost of ML at scale is not the models - it is the duplicated feature engineering work every team does independently. What a feature store is and whether you actually need one.