Insights & Updates
The latest thinking on Data Strategy, AI Engineering, and Modern Analytics architectures.
Understanding the "Harness" in Agentic Coding Tools: Claude Code vs Codex
The model is the reasoning engine. The harness is everything that makes it actually useful in your codebase — tools, context management, orchestration, memory, safety. Here's what a harness does, what separates a good one from a fragile one, and how Claude Code and Codex compare.
Digital Twins: Simulating Scenarios Before They Cost You an Incident
Strip away the hype and a digital twin is a disciplined answer to "what happens if?"—a live-data model that lets you rehearse the dangerous and the expensive safely. Here's what that actually means for data pipelines and software systems.
Cutting Claude Code Token Costs by Offloading to Open-Weight Models — and the Trust Problem Nobody Puts on the Invoice
Routing routine agentic coding work to a cheap open-weight model like GLM-5.2 can slash your token bill—but the real cost is a trust risk that never shows up on the invoice.
The Benefits of the Microsoft Fabric Well-Architected Framework
How applying WAF's five pillars—reliability, security, cost optimization, operational excellence, and performance efficiency—transforms your Microsoft Fabric data platform from functional to enterprise-ready.
The Field-Level Lineage Gap in Microsoft Fabric (And How We're Closing It)
Someone asks where a number in the report actually comes from, and tracing it requires navigating six notebooks, three warehouse views, and a semantic model. Here's how we're solving that.
Forcing a Metadata Sync in Microsoft Fabric Lakehouse
Having written before about the frustrations around records arriving late in the Fabric Lakehouse, I revisited the code today and discovered Microsoft's official REST API solution.
Delta Live Tables in Databricks… meet Materialized Lakehouse Views in Microsoft Fabric
If you've built Delta Live Tables pipelines in Databricks, the newly released Materialized Lakehouse Views in Microsoft Fabric will feel familiar. Here's how they compare.
Where are you on your Data Maturity Journey?
And where do you think your competitors are on their journey? If you are lucky, you may have an idea. But many companies simply have no idea how they compare.
Is CI/CD for Fabric finally completely there?
For years, data engineers and BI developers have been asking: when will Microsoft Fabric have proper CI/CD pipelines? The answer: it's close enough now.
From BI to HI to AI - Why Human Intelligence is the Real Driver of Smart Business
Over the past decade, Business Intelligence (BI) transformed how we see our data. Now, Artificial Intelligence (AI) is transforming what we can predict. But there's a missing link: Human Intelligence (HI).
The Rise of Agentic AI in Data Engineering
Why traditional ETL pipelines are being replaced by autonomous agents that heal themselves.
RAG vs Fine-Tuning: Making the Right Choice
When to inject context and when to retrain the model. A strategic guide for CTOs.