The next article initially appeared on Medium and is being republished right here with the writer’s permission.
There’s a faux confidence you’ll be able to carry round once you’re studying a brand new know-how. You watch a couple of movies, skim some docs, get a toy instance working, and inform your self, “Yeah, I’ve bought this.” I’ve achieved that. It by no means lasts. A troublesome lesson usually accompanies the one expertise that issues.
You be taught by way of failure—falling flat in your face, wanting on the mess, and determining why it broke. Something that feels too straightforward? It in all probability was, and also you didn’t exit the method with something value studying.
Ask About Failure: Failure === Expertise
Once I’m hiring somebody who claims relational database experience, I ask a “trick” query:
Inform me in regards to the worst database schema you ever created. What did it educate you to keep away from?
It’s not likely a trick. Anybody who’s been knee‑deep in relational databases is aware of there’s no good schema. There are competing use circumstances that continuously pull in opposition to one another. You design for transaction workloads, however inevitably, somebody tries to make use of it for reporting, then everybody wonders why queries crawl. One other developer on the group inadvertently optimizes the schema (normally years later) for the reporting use case solely to make the transactional workload unworkable.
The right reply normally feels like:
We constructed for transactional throughput—one of many founders of the corporate thought MySQL was a database, which was our first mistake. The enterprise then used it for reporting functions. The system modified arms a number of occasions over the course of a number of years. Joins grew to become gnarly, indices didn’t match the entry patterns, and nightly jobs began interfering with consumer site visitors. We needed to break up learn replicas, ultimately introduce a warehouse, and after 5–6 years, we ended up simplifying the transactions and transferring them over to Cassandra.
That’s an individual who has lived the trade-offs. They’ve skilled a drawn-out existential failure associated to operating a database. Whereas they may not know tips on how to remedy a number of the foolish logic questions which can be more and more fashionable in job interviews, that is the kind of expertise that carries much more weight with me.
The Schema That Practically Broke Me
I as soon as shipped a transactional schema that seemed high quality on paper: normalized, neat, every little thing in its correct place.
Then analytics confirmed up with “simply a few fast dashboards.” Subsequent factor you already know, my fairly 3NF mannequin, now linked to each elementary classroom in America, was getting used like a million-row Excel spreadsheet to summarize an accounting report. For a couple of months, it was high quality till it wasn’t, and the database had made a sluggish‑movement faceplant as a result of it was spending 80% of its time updating an index. It wasn’t as if I may repair something, as a result of that may imply a number of days of downtime coupled with a rewrite for a mission whose contract was nearly up.
And the way have been we making an attempt to repair it? When you’ve been on this scenario, you’ll perceive that what I’m about to write down is the signal that you’ve reached a brand new stage of determined failure. As an alternative of contemplating a rational strategy to reform the schema or separating what had develop into a “web-scale” workload in 2007 from a NoSQL database, we have been making an attempt to determine tips on how to buy sooner arduous drives with larger IOPS.
I realized a variety of issues:
- I realized that upgrading {hardware} (shopping for a sooner machine or dropping one million {dollars} on arduous drives) will solely delay your disaster. The true repair is unavoidable—large horizontal scaling is incompatible with relational databases.
- I realized the which means of “question plan from hell.” We band‑aided it with materialized views and browse replicas. Then we did what we should always’ve achieved from day one: arrange an precise reporting path.
- In case you are having to optimize for a question plan each week? Your database is sending you an vital sign, which it’s best to translate to, “It’s time to start out on the lookout for another.”
Lesson burned in: Design for the use case you even have, not the one you hope to have—and assume the use case will change.
What Does This Need to Do with Cursor and Copilot?
I’m seeing lots of people writing on LinkedIn and different websites about how superb vibe coding is. These celebratory posts reveal extra in regards to the folks posting them than they notice, as they not often acknowledge the fact of the method—it’s not all enjoyable and video games. Whereas it’s astonishing how a lot progress one could make in a day or per week, these of us who’re truly utilizing these instruments to write down code are the primary to let you know that we’re studying a variety of troublesome classes.
It’s not “straightforward.” There’s nothing “vibey” in regards to the course of, and if you’re doing it proper, you might be beginning to use curse phrases in your prompts. For instance, a few of my prompts in response to a Cursor Agent yesterday have been: “You’ve got to be kidding me, I’ve a rule that said that I by no means needed you to do this, and also you simply ignored it?”
At any time when I see folks get excited in regards to the newest, best fad factor that’s altering the world, I’m additionally the primary to note that perhaps they aren’t utilizing all of it. In the event that they have been, they’d perceive that it’s not as “straightforward” as they’re reporting.
The failure muscle you construct with databases is similar one you want with AI coding instruments. You may’t tiptoe in. You need to push till one thing breaks. Then you determine tips on how to strategy a brand new know-how as knowledgeable.
- Ask an agent to refactor one file—nice.
- Ask it to coordinate adjustments throughout 20 information, rethink error dealing with, and maintain exams passing—now we’re studying.
- Watch the place it stumbles, and be taught to body the work so it will probably succeed subsequent time.
- Spend a whole weekend on a “wild goose chase” as a result of your agentic coder determined to disregard your Cursor guidelines fully. ← That is costly, nevertheless it’s the way you be taught.
The trick isn’t avoiding failure. It’s failing in a managed, reversible method.
The Meta Lesson
When you’ve by no means damaged it, you don’t actually understand it. That is true for coding, budgeting, managing, cooking, and snowboarding. When you haven’t failed, you don’t understand it. And most people speaking about “vibe coding” haven’t.
The folks I belief most as engineers can inform me why one thing failed and how they adjusted their strategy because of this. That’s all the sport with AI coding instruments. The sooner you’ll be able to run the loop—strive → break → examine → refine—the higher you get.
