The Expert Trap: When Knowing the Pattern Stops You Listening
Hired for my Azure expertise on a Fabric migration, I found the client's engineers knew more than I did, and learned why that was exactly the point.
Thoughts on data engineering, cloud architecture, and technology. 2205 articles and counting.
Showing 24 of 2205 posts — page 6 of 92
Hired for my Azure expertise on a Fabric migration, I found the client's engineers knew more than I did, and learned why that was exactly the point.
Azure OpenAI caches prompt prefixes automatically, yet many apps see few cache hits. How to measure the hit rate and fix the prompt shapes that break it.
What Microsoft Agent Framework for .NET really is in February 2026: agents, sessions, tool approvals and handoffs, plus where it fits before it reaches GA.
Agent Framework is still in preview and Semantic Kernel is in fix-and-stabilise mode: how to decide whether to stay, bridge, or rewrite in February 2026.
A production alert on date night taught me my wife doesn't need grand gestures. She needs my attention, and my phone was taking it.
Run AI monitoring as an operating rhythm: four layers with owners, what pages versus what waits, and tagging every change so regressions are traceable.
Why I stopped saying yes to every consulting project, the three filters I now run each offer through, and how to decline without burning the relationship.
Most agent designs only need autonomy in one step. How to build the workflow around it with Agent Framework executors, switch-case edges and loop caps.
Forty-one days of daily blogging taught me I write to find ideas, not share them, and that cutting half the draft is where the thinking happens.
What consolidating Azure Data Factory, Databricks, Azure SQL and Power BI onto Microsoft Fabric delivered in production, and where the seams still show.
When Andriel asked what I do at work, I wrote down what I want my kids to learn first: critical thinking, privacy, boundaries and making things.
An LLM judge has its own biases: how to write pass/fail criteria, measure its kappa and false-pass rate against human labels, and know when to skip it.
The first family day with phones, laptops and tablets put away: the itch in the first hour, the fake-urgent work thoughts, and why we're doing it again.
Eight years of data pipelines taught me to plan for reruns, late data and schema drift. The habits that matter, with Delta MERGE and Fabric examples.
With the Azure OpenAI v1 API one OpenAI client works on both, so the choice rests on data processing location, identity, networking, capacity and model timing.
Why scattered 30-minute meetings can wipe out a day of deep work, what the research actually says about interruptions, and the schedule rules I now use.
Azure AI Search knowledge bases now plan queries for you, in preview. When to hand over the RAG pipeline, when to keep building it, and how to test both.
February plan after 31 days of daily writing: keep the habit, take Sundays off tech, narrow to two focus areas, and stop running on five to six hours of sleep.
The final tally on January: 31 posts in 31 days, one working Saturday, three habit experiments that failed, and the short list I'm taking into February.
A minimal GitHub Actions to App Service container pipeline for small apps, and the specific signals that tell you when slots, canaries or flags are worth it.
How I moved from cloud and data engineering into AI work without a maths degree: what I learnt in which order, what I skipped, and where engineers go wrong.
The AI security failures that actually cost money are leaked keys, unbounded token spend and sensitive data in logs. Fix those before prompt defences.
A mid-course report card on four January commitments: weekend boundaries, being present, daily writing and AI focus, plus three small goals for February.
A practitioner's comparison of Azure DevOps and GitHub as complete platforms in early 2026: planning, repos, CI/CD, security, cost and when to run both.