coding is right here to remain. Instruments like Cursor, V0, and Lovable have dramatically lowered the barrier to entry — constructing dashboards, pipelines, or complete apps can now be executed in a fraction of the time.
I exploit these instruments day by day, and so they’ve undoubtedly made me quicker. However because the codebase will get extra advanced, the tradeoffs turn out to be clear: cryptic bugs, tangled logic, and hours misplaced debugging code I didn’t actually perceive.
AI instruments are nice — particularly for newcomers — however they arrive with a quiet value. The extra you let AI carry the load, the less probabilities you need to sharpen your instincts that come from wrestling with complexity.
Sure, AI will velocity up your workflow, however you’ll additionally skip the formative steps the place technical knowledge is earned.
“Vibe coding” — shortly cobbling collectively code with minimal planning — is nice for demos or experiments. However for deeper technical progress or constructing methods with significant complexity, vibe coding isn’t sufficient. This trending Reddit publish sums it up completely: left unchecked, vibe coding creates extra issues than it solves.
On this piece, I’ll present you the way to use AI-assisted instruments extra properly — and why contributing to Open Supply could be probably the most underrated strategy to actually stage up your technical abilities.
My expertise vibe coding with Cursor
Like many builders, I switched from VS Code (with GitHub Copilot) to Cursor and am presently subscribed to Cursor’s Professional plan ($20/month).
The function I depend on most is Cursor’s built-in AI chat, which lets me instantly work together with my complete codebase. Its agent can shortly grep via a number of recordsdata and even deal with photographs - extraordinarily helpful when navigating massive, unfamiliar repos. It additionally spots linter errors and auto-corrects them whereas instantly modifying recordsdata.
Initially, Cursor dramatically boosted my productiveness, particularly for easier duties. It felt highly effective, nearly magical. However as issues bought advanced, I observed some cracks. Cursor would generally generate spaghetti code, combine up equally named recordsdata throughout directories, and infrequently battle to comply with intricate logic flows.
Vibe coding can get you hundreds of strains of code in minutes — however with no sturdy psychological mannequin of what you’re constructing, you threat ending up with bloated, over-engineered methods.
Cursor does an honest job narrowing down the search house when debugging. However letting it make unchecked edits does introduce much more bugs than it solves.
Past the standard recommendation to “write higher prompts,” one technique I’ve discovered particularly useful is telling Cursor NOT to make direct edits. (It’s surprisingly obedient about this!)
As an alternative, I explicitly ask it to counsel adjustments first within the chat interface. Then, I might evaluation every suggestion, determine which edits made sense, and apply them selectively — both manually or via Cursor. In contrast to ChatGPT, Cursor’s greatest energy is its contextual consciousness of your entire codebase and its means to parse via prolonged recordsdata (over 5k strains of code) by processing them in manageable chunks.
Contributing to open supply
So, how do you get technically stronger? Two methods stand out: facet tasks and open supply contributions.
Facet tasks are nice for exploring new tech or diving deep into one thing you’re passionate or interested in. Surprise how AI brokers work or interested in MCP? Simply constructing a easy weekend challenge teaches you excess of hours of tutorials or documentation. Due to open-source, instruments and assets are freely accessible, leveling the taking part in subject for everybody.
However solo tasks have downsides. It’s straightforward to lose motivation — lots of my very own facet tasks by no means noticed the sunshine of day.
Plus, you’ll find your self in an echo chamber: your code works, however you’re undecided if it’s following finest practices or business requirements. In case you’re early in your profession and lack mentorship, how are you aware if you happen to’re even heading in the right direction?
That is precisely the place open supply fills the hole. Open supply tasks aren’t only for coding wizards; they’re for everybody. Your favorite libraries like Pandas, Matplotlib, TensorFlow, and Keras rely closely on neighborhood involvement.
Why hassle contributing?
Open supply allows you to make an actual impression utilized by hundreds of builders — not simply toy tasks no one sees. You’ll turn out to be proficient with model management (whats up, GitHub!), sharpen your abilities navigating advanced codebases, decide up finest practices, and construct a community of people that can vouch for you when it issues.
There are profession advantages too. It’ll add to your portfolio and private model, and also you’ll ramp up quicker when becoming a member of new groups.
However, contribute for the appropriate causes. In case your solely motivation is touchdown a job, DON’T contribute! Open supply just isn’t a ticket to get a job — it requires real curiosity and dedication. It exhibits you’ve a ardour to construct, and for a lot of startups that start from open supply tasks, that’s how they discover their first hires.
Choosing an open supply challenge that you just care about
Beginning out can appear daunting. Many common repos have huge codebases, probably outdated documentation, or a whole bunch of unclear points. So how do you decide?
First up, decide a challenge you genuinely care about. This would possibly sound apparent, however it’s essential — and underrated.
Select one thing you truly use, whether or not at work or in a facet challenge. Leaping into an unfamiliar challenge with unfamiliar tech is just overwhelming, and also you’ll lose motivation quick.
Personally, I’m each a consumer and an enormous fan of PostHog — the product analytics platform constructed particularly for builders — so I began contributing there. Their docs had been complete and well-structured, which made it an superior place to begin. (And no, they didn’t pay me to say this!)
What to contribute?
There are a ton of issues you are able to do. Right here’s an method that I discovered useful.
- Discover a function you want or enhance one thing you employ.
Narrowing down contributions to options you genuinely care about provides readability and motivation. The most effective code comes from fixing issues you personally face. - Arrange your native surroundings.
Fork the challenge, clone it regionally, and get it working. Perceive the place logs are and the way to take a look at adjustments. Get a grasp on the challenge’s primary construction and coding model. - Begin small and study by doing
Many repos tag beginner-friendly points (like “good-first-issue”). Choose these to begin. Perceive and replicate the bug; don’t hesitate to remark if you happen to’re caught. If you open a PR, guarantee your adjustments cross all linting and checks.
Studying to navigate the codebase is important. You don’t have to learn each line — that’s virtually unimaginable. After greedy the high-level construction, dive in. Begin small to get comfy with the construct, deployment, and PR evaluation course of. Write clear commit messages and PR descriptions. Verify not too long ago merged PRs to see profitable examples or insightful discussions.
Wrapping up
Contributing to open supply takes persistence — common repos are enormous, and studying takes time. Changing into a constant, useful contributor takes no less than a couple of months, so don’t get discouraged by preliminary setbacks. In case your PR is rejected otherwise you get caught on a difficult bug, that’s completely regular — it’s all a part of the educational course of.
In case you’re new to open supply and wish to chat, be happy to attach. Whereas I didn’t dive deeply into technical particulars right here (a fast Google or ChatGPT search can information you there), I hope this provides you the big-picture perspective to get began. Open supply has been rewarding for me — and I hope it will likely be for you too.
See you within the subsequent article 🙂