Builders are doing unbelievable issues with AI. Instruments like Copilot, ChatGPT, and Claude have quickly change into indispensable for builders, providing unprecedented velocity and effectivity in duties like writing code, debugging difficult conduct, producing assessments, and exploring unfamiliar libraries and frameworks. When it really works, it’s efficient, and it feels extremely satisfying.
However in the event you’ve spent any actual time coding with AI, you’ve in all probability hit some extent the place issues stall. You retain refining your immediate and adjusting your strategy, however the mannequin retains producing the identical form of reply, simply phrased somewhat otherwise every time, and returning slight variations on the identical incomplete resolution. It feels shut, however it’s not getting there. And worse, it’s not clear find out how to get again on observe.
That second is acquainted to lots of people making an attempt to use AI in actual work. It’s what my latest speak at O’Reilly’s AI Codecon occasion was all about.
During the last two years, whereas engaged on the most recent version of Head First C#, I’ve been creating a brand new form of studying path, one which helps builders get higher at each coding and utilizing AI. I name it Sens-AI, and it got here out of one thing I saved seeing:
There’s a studying hole with AI that’s creating actual challenges for people who find themselves nonetheless constructing their improvement expertise.
My latest O’Reilly Radar article “Bridging the AI Studying Hole” checked out what occurs when builders attempt to be taught AI and coding on the similar time. It’s not only a tooling downside—it’s a considering downside. A number of builders are figuring issues out by trial and error, and it grew to become clear to me that they wanted a greater strategy to transfer from improvising to truly fixing issues.
From Vibe Coding to Downside Fixing
Ask builders how they use AI, and plenty of will describe a form of improvisational prompting technique: Give the mannequin a activity, see what it returns, and nudge it towards one thing higher. It may be an efficient strategy as a result of it’s quick, fluid, and nearly easy when it really works.
That sample is widespread sufficient to have a reputation: vibe coding. It’s an excellent start line, and it really works as a result of it attracts on actual immediate engineering fundamentals—iterating, reacting to output, and refining based mostly on suggestions. However when one thing breaks, the code doesn’t behave as anticipated, or the AI retains rehashing the identical unhelpful solutions, it’s not at all times clear what to strive subsequent. That’s when vibe coding begins to crumble.
Senior builders have a tendency to select up AI extra shortly than junior ones, however that’s not a hard-and-fast rule. I’ve seen brand-new builders decide it up shortly, and I’ve seen skilled ones get caught. The distinction is in what they do subsequent. The individuals who succeed with AI are likely to cease and rethink: They determine what’s going improper, step again to take a look at the issue, and reframe their immediate to provide the mannequin one thing higher to work with.
The Sens-AI Framework
As I began working extra carefully with builders who have been utilizing AI instruments to attempt to discover methods to assist them ramp up extra simply, I paid consideration to the place they have been getting caught, and I began noticing that the sample of an AI rehashing the identical “nearly there” ideas saved developing in coaching classes and actual initiatives. I noticed it occur in my very own work too. At first it felt like a bizarre quirk within the mannequin’s conduct, however over time I spotted it was a sign: The AI had used up the context I’d given it. The sign tells us that we’d like a greater understanding of the issue, so we can provide the mannequin the knowledge it’s lacking. That realization was a turning level. As soon as I began being attentive to these breakdown moments, I started to see the identical root trigger throughout many builders’ experiences: not a flaw within the instruments however an absence of framing, context, or understanding that the AI couldn’t provide by itself.
Over time—and after numerous testing, iteration, and suggestions from builders—I distilled the core of the Sens-AI studying path into 5 particular habits. They got here straight from watching the place learners received caught, what sorts of questions they requested, and what helped them transfer ahead. These habits type a framework that’s the mental basis behind how Head First C# teaches builders to work with AI:
- Context: Being attentive to what info you provide to the mannequin, making an attempt to determine what else it must know, and supplying it clearly. This consists of code, feedback, construction, intent, and the rest that helps the mannequin perceive what you’re making an attempt to do.
- Analysis: Actively utilizing AI and exterior sources to deepen your personal understanding of the issue. This implies operating examples, consulting documentation, and checking references to confirm what’s actually happening.
- Downside framing: Utilizing the knowledge you’ve gathered to outline the issue extra clearly so the mannequin can reply extra usefully. This entails digging deeper into the issue you’re making an attempt to resolve, recognizing what the AI nonetheless must find out about it, and shaping your immediate to steer it in a extra productive route—and going again to do extra analysis whenever you understand that it wants extra context.
- Refining: Iterating your prompts intentionally. This isn’t about random tweaks; it’s about making focused adjustments based mostly on what the mannequin received proper and what it missed, and utilizing these outcomes to information the following step.
- Vital considering: Judging the standard of AI output fairly than simply merely accepting it. Does the suggestion make sense? Is it appropriate, related, believable? This behavior is particularly essential as a result of it helps builders keep away from the lure of trusting confident-sounding solutions that don’t truly work.
These habits let builders get extra out of AI whereas protecting management over the route of their work.
From Caught to Solved: Getting Higher Outcomes from AI
I’ve watched numerous builders use instruments like Copilot and ChatGPT—throughout coaching classes, in hands-on workouts, and after they’ve requested me straight for assist. What stood out to me was how typically they assumed the AI had finished a foul job. In actuality, the immediate simply didn’t embrace the knowledge the mannequin wanted to resolve the issue. Nobody had proven them find out how to provide the suitable context. That’s what the 5 Sens-AI habits are designed to deal with: not by handing builders a guidelines however by serving to them construct a psychological mannequin for find out how to work with AI extra successfully.
In my AI Codecon speak, I shared a narrative about my colleague Luis, a really skilled developer with over three many years of coding expertise. He’s a seasoned engineer and a sophisticated AI consumer who builds content material for coaching different builders, works with giant language fashions straight, makes use of refined prompting methods, and has constructed AI-based evaluation instruments.
Luis was constructing a desktop wrapper for a React app utilizing Tauri, a Rust-based toolkit. He pulled in each Copilot and ChatGPT, cross-checking output, exploring alternate options, and making an attempt totally different approaches. However the code nonetheless wasn’t working.
Every AI suggestion appeared to repair a part of the issue however break one other half. The mannequin saved providing barely totally different variations of the identical incomplete resolution, by no means fairly resolving the difficulty. For some time, he vibe-coded by way of it, adjusting the immediate and making an attempt once more to see if a small nudge would assist, however the solutions saved circling the identical spot. Finally, he realized the AI had run out of context and adjusted his strategy. He stepped again, did some targeted analysis to higher perceive what the AI was making an attempt (and failing) to do, and utilized the identical habits I emphasize within the Sens-AI framework.
That shift modified the result. As soon as he understood the sample the AI was making an attempt to make use of, he might information it. He reframed his immediate, added extra context, and eventually began getting ideas that labored. The ideas solely began working as soon as Luis gave the mannequin the lacking items it wanted to make sense of the issue.
Making use of the Sens-AI Framework: A Actual-World Instance
Earlier than I developed the Sens-AI framework, I bumped into an issue that later grew to become a textbook case for it. I used to be curious whether or not COBOL, a decades-old language developed for mainframes that I had by no means used earlier than however needed to be taught extra about, might deal with the essential mechanics of an interactive recreation. So I did some experimental vibe coding to construct a easy terminal app that will let the consumer transfer an asterisk across the display screen utilizing the W/A/S/D keys. It was a bizarre little facet mission—I simply needed to see if I might make COBOL do one thing it was by no means actually meant for, and be taught one thing about it alongside the way in which.
The preliminary AI-generated code compiled and ran simply high quality, and at first I made some progress. I used to be capable of get it to clear the display screen, draw the asterisk in the suitable place, deal with uncooked keyboard enter that didn’t require the consumer to press Enter, and get previous some preliminary bugs that prompted numerous flickering.
However as soon as I hit a extra delicate bug—the place ANSI escape codes like ";10H" have been printing actually as an alternative of controlling the cursor—ChatGPT received caught. I’d describe the issue, and it will generate a barely totally different model of the identical reply every time. One suggestion used totally different variable names. One other modified the order of operations. A number of tried to reformat the STRING assertion. However none of them addressed the basis trigger.
The sample was at all times the identical: slight code rewrites that seemed believable however didn’t truly change the conduct. That’s what a rehash loop appears like. The AI wasn’t giving me worse solutions—it was simply circling, caught on the identical conceptual concept. So I did what many builders do: I assumed the AI simply couldn’t reply my query and moved on to a different downside.
On the time, I didn’t acknowledge the rehash loop for what it was. I assumed ChatGPT simply didn’t know the reply and gave up. However revisiting the mission after creating the Sens-AI framework, I noticed the entire alternate in a brand new mild. The rehash loop was a sign that the AI wanted extra context. It received caught as a result of I hadn’t informed it what it wanted to know.
After I began engaged on the framework, I remembered this previous failure and thought it’d be an ideal check case. Now I had a set of steps that I might observe:
- First, I acknowledged that the AI had run out of context. The mannequin wasn’t failing randomly—it was repeating itself as a result of it didn’t perceive what I used to be asking it to do.
- Subsequent, I did some focused analysis. I brushed up on ANSI escape codes and began studying the AI’s earlier explanations extra rigorously. That’s once I seen a element I’d skimmed previous the primary time whereas vibe coding: After I went again by way of the AI rationalization of the code that it generated, I noticed that the
PIC ZZCOBOL syntax defines a numeric-edited area. I suspected that would probably trigger it to introduce main areas into strings and puzzled if that would break an escape sequence. - Then I reframed the issue. I opened a brand new chat and defined what I used to be making an attempt to construct, what I used to be seeing, and what I suspected. I informed the AI I’d seen it was circling the identical resolution and handled that as a sign that we have been lacking one thing basic. I additionally informed it that I’d finished some analysis and had three leads I suspected have been associated: how COBOL shows a number of objects in sequence, how terminal escape codes must be formatted, and the way spacing in numeric fields could be corrupting the output. The immediate didn’t present solutions; it simply gave some potential analysis areas for the AI to research. That gave it what it wanted to seek out the extra context it wanted to interrupt out of the rehash loop.
- As soon as the mannequin was unstuck, I refined my immediate. I requested follow-up inquiries to make clear precisely what the output ought to appear to be and find out how to assemble the strings extra reliably. I wasn’t simply on the lookout for a repair—I used to be guiding the mannequin towards a greater strategy.
- And most of all, I used important considering. I learn the solutions carefully, in contrast them to what I already knew, and determined what to strive based mostly on what truly made sense. The reason checked out. I carried out the repair, and this system labored.
As soon as I took the time to know the issue—and did simply sufficient analysis to provide the AI a number of hints about what context it was lacking—I used to be capable of write a immediate that broke ChatGPT out of the rehash loop, and it generated code that did precisely what I wanted. The generated code for the working COBOL app is accessible in this GitHub GIST.
Why These Habits Matter for New Builders
I constructed the Sens-AI studying path in Head First C# across the 5 habits within the framework. These habits aren’t checklists, scripts, or hard-and-fast guidelines. They’re methods of considering that assist individuals use AI extra productively—and so they don’t require years of expertise. I’ve seen new builders decide them up shortly, typically quicker than seasoned builders who didn’t understand they have been caught in shallow prompting loops.
The important thing perception into these habits got here to me once I was updating the coding workouts in the newest version of Head First C#. I check the workouts utilizing AI by pasting the directions and starter code into instruments like ChatGPT and Copilot. In the event that they produce the proper resolution, which means I’ve given the mannequin sufficient info to resolve it—which suggests I’ve given readers sufficient info too. But when it fails to resolve the issue, one thing’s lacking from the train directions.
The method of utilizing AI to check the workouts within the ebook jogged my memory of an issue I bumped into within the first version, again in 2007. One train saved tripping individuals up, and after studying numerous suggestions, I spotted the issue: I hadn’t given readers all the knowledge they wanted to resolve it. That helped join the dots for me. The AI struggles with some coding issues for a similar motive the learners have been fighting that train—as a result of the context wasn’t there. Writing coding train and writing immediate each rely on understanding what the opposite facet must make sense of the issue.
That have helped me understand that to make builders profitable with AI, we have to do extra than simply train the fundamentals of immediate engineering. We have to explicitly instill these considering habits and provides builders a strategy to construct them alongside their core coding expertise. If we wish builders to succeed, we are able to’t simply inform them to “immediate higher.” We have to present them find out how to assume with AI.
The place We Go from Right here
If AI actually is altering how we write software program—and I imagine it’s—then we have to change how we train it. We’ve made it straightforward to provide individuals entry to the instruments. The tougher half helps them develop the habits and judgment to make use of them nicely, particularly when issues go improper. That’s not simply an schooling downside; it’s additionally a design downside, a documentation downside, and a tooling downside. Sens-AI is one reply, however it’s only the start. We nonetheless want clearer examples and higher methods to information, debug, and refine the mannequin’s output. If we train builders find out how to assume with AI, we may also help them change into not simply code mills however considerate engineers who perceive what their code is doing and why it issues.
