DEV Community: member_fc281ffe

[Boost]

member_fc281ffe — Sun, 22 Feb 2026 14:55:20 +0000

The Internet Shows What We Do. It Hides How We Feel.

Raven ・ Feb 22

#discuss #mentalhealth #showdev #sideprojects

[Boost]

member_fc281ffe — Sun, 22 Feb 2026 14:49:20 +0000

I Built a Tiny MCP That Understands Your Code and Saves 70% Tokens

Linghua Jin ・ Feb 22

#ai #programming #productivity #opensource

[Boost]

member_fc281ffe — Sun, 22 Feb 2026 14:43:17 +0000

StackOverflow - was it worth it?

Nikola Brežnjak ・ Feb 21

#stackoverflow #career

[Boost]

member_fc281ffe — Sun, 22 Feb 2026 14:38:07 +0000

The Internet Shows What We Do. It Hides How We Feel.

Raven ・ Feb 22

#discuss #mentalhealth #showdev #sideprojects

[Boost]

member_fc281ffe — Sun, 22 Feb 2026 14:32:29 +0000

Promotions Don't Go to the Best Coders (Here's Why)

Mark ・ Feb 22

#career #webdev #beginners #programming

[Boost]

member_fc281ffe — Sun, 22 Feb 2026 14:26:47 +0000

I Built a Tiny MCP That Understands Your Code and Saves 70% Tokens

Linghua Jin ・ Feb 22

#ai #programming #productivity #opensource

[Boost]

member_fc281ffe — Sun, 22 Feb 2026 14:21:43 +0000

Teaching a Robot to Play a Toddler Game: VLAs, Gemini 3 Flash, and First Orchard

Paul Ruiz for Google AI ・ Feb 21

#gemini #robotics #ai #python

My Tech Interview Prep System (No Leetcode Grinding Required)

member_fc281ffe — Sun, 22 Feb 2026 10:20:30 +0000

I used to prep for tech interviews the way most people do: panic-grind Leetcode for two weeks, memorize answers to "tell me about yourself," and hope for the best.

It worked sometimes. Mostly it didn't. And when it didn't, I had no idea what went wrong.

So I built a system. Nothing fancy — just a repeatable process that takes the randomness out of interview prep. I've refined it through my own prep process and it's made a real difference. Here's the whole thing.

Step 1: Know what they're actually screening for

Before you prep anything, figure out what the company cares about. This sounds obvious, but most people skip it.

Read the job posting carefully. Not the requirements list — the description of what you'll actually be doing. Then check Glassdoor or Blind for interview experiences at that company. You're looking for patterns: Do they do system design? Live coding? Behavioral deep-dives?

Tailoring your prep to the actual interview format saves you from wasting time on stuff that won't come up.

Step 2: Build your story bank

Most behavioral questions are variations of five themes: conflict, failure, leadership, ambiguity, and impact. Write one strong story for each. Use the STAR format (Situation, Task, Action, Result), but keep it conversational. Nobody wants to hear a rehearsed monologue.

I keep mine in a simple doc and update it after every project. When interview time comes, I just review and adapt.

Step 3: Practice system design out loud

The system design round trips up experienced engineers more than anything else. Not because they can't design systems — they do it every day. It's because they can't explain their design process while doing it.

Practice talking through your decisions. "I'm choosing a message queue here because..." Force yourself to articulate the tradeoffs. Get a friend to ask "why not X?" questions. The goal isn't to memorize architectures. It's to get comfortable thinking out loud.

Step 4: Do two targeted coding problems per day

Not twenty. Two. Pick problems related to the patterns that company tests for. Spend 25 minutes max per problem. If you can't solve it in 25 minutes, read the solution, understand the pattern, and move on.

The point isn't to solve every problem. It's to recognize patterns quickly so you're not starting from zero in the actual interview.

Step 5: Run mock interviews

This is the highest-ROI thing you can do and the thing everyone skips. Find someone — a friend, a Discord community, whatever — and do a live mock interview. The gap between "I know this" and "I can perform this under pressure" is enormous.

The meta-lesson

Interview prep isn't about knowing more. It's about presenting what you already know under artificial pressure. A system helps because it removes the "what should I even work on today?" decision fatigue.

Build a system. Follow it. Trust it.

I put together a more detailed version of this as a guide — including the actual story bank template, system design frameworks, and the question patterns I track. You can grab it here: The Interview Playbook. And if you want to make sure your resume actually gets through ATS filters first, I have a free ATS checklist too.

You're Not Behind on AI. Here's a Practical Starting Point.

member_fc281ffe — Sun, 22 Feb 2026 10:11:03 +0000

I keep seeing the same post in different words: "I'm a senior dev and I feel lost on AI."

I get it. Every week there's a new model, a new framework, a new thing you're apparently supposed to know. The discourse makes it sound like if you haven't rebuilt your entire workflow around AI by now, you're already obsolete.

That's not true. Let me offer a different perspective.

Most of the noise doesn't matter

Here's what I mean. In the last month, I've seen announcements for probably 40 new AI tools. I use three of them regularly. Three. The other 37 either duplicate something I already have, solve a problem I don't have, or are so early they break more than they help.

The people who look like they're "ahead" on AI aren't tracking every release. They picked one or two tools, used them daily, and got good at those specific tools. That's it.

A starting point that actually works

If you're a working developer and you want to start using AI without drowning in hype, here's what I'd suggest:

Week 1: Use AI for code explanation only. Take a piece of code you didn't write — maybe from a library, maybe from a coworker. Paste it into any LLM and ask it to explain what's happening. Don't ask it to write code yet. Just use it as a reading companion.

Week 2: Use it for tests. Write your implementation as normal. Then ask AI to generate unit tests for it. Read every test it writes. You'll learn a lot about edge cases you missed, and you'll also learn where AI makes bad assumptions.

Week 3: Use it for first drafts. Now try asking AI to write a small function. A utility. Something contained. Compare its approach to what you would've written. Notice the differences.

Week 4: Decide what stays. By now you'll have opinions. You'll know which parts of your workflow AI actually improves and which parts it slows down. Keep the good parts. Drop the rest.

What I wish someone told me earlier

AI doesn't replace the judgment you've built over years of writing software. It doesn't know your codebase, your team's conventions, or why you made that weird architectural decision three months ago.

What it does is compress the boring parts. The parts that eat your time but not your brain. That's genuinely useful. It's just not the revolution people are selling.

If you've been coding for 20+ years, you have something AI literally cannot replicate: context. Real-world context from shipping real things. That's your edge. AI is just a tool that makes that edge sharper.

I write about the practical side of using AI as a developer — what works, what doesn't, and what's just marketing. More on my Dev.to profile.

I Let AI Write My Code for a Week. Here's What It Got Wrong Every Time.

member_fc281ffe — Sun, 22 Feb 2026 10:10:42 +0000

Last week I ran a small experiment. I let AI coding assistants handle as much of my daily work as possible — Cursor, Copilot, Claude in the terminal, the works. Not to prove a point. I genuinely wanted to know where the ceiling is right now.

Here's my honest take after five days.

What it nailed

Boilerplate. Setting up Express routes, writing test skeletons, generating TypeScript interfaces from JSON — all that stuff that eats 20 minutes but requires zero creativity. AI handles it faster than I can type.

Explaining unfamiliar code. I inherited a gnarly regex-heavy config parser last month. Pasting it into Claude and asking "what does this do, line by line" saved me an hour of squinting.

Refactoring suggestions. "Extract this into a utility function" type stuff. Not always right, but it got me thinking in the right direction more often than not.

What it got wrong — every single time

Deleting things it shouldn't. I asked Cursor to clean up a utility file. It removed every comment. Not refactored — deleted. The comments were the only documentation that file had. I've seen this pattern repeatedly: AI assistants treat comments as noise.

Making up APIs that don't exist. I lost count of how many times it generated code calling methods that were never part of the library. It writes with such confidence that you don't question it until runtime. And if you're not running tests immediately, that bug hides for days.

Ignoring the surrounding codebase. This is the big one. AI writes code that works in isolation, but doesn't match the patterns already in your project. Different error handling style. Different naming conventions. It's like hiring a contractor who does good work but never reads your team's style guide.

The pattern I actually settled into

By Thursday, I'd stopped asking AI to "write this feature." Instead I'd:

Write the rough implementation myself (15 min)
Ask AI to review it for edge cases (2 min)
Ask AI to write the tests (5 min)
Actually read the tests and fix the ones that test the wrong thing (10 min)

Total: 32 minutes instead of my usual 45, but with better test coverage. Not revolutionary. Just... better.

The uncomfortable truth

AI coding tools right now are like a very fast junior dev who never sleeps but also never asks clarifying questions. The output looks professional until you look closely.

The people getting the most out of these tools aren't the ones outsourcing their thinking. They're the ones who already know what good code looks like and use AI to get there faster.

If you're still figuring out your approach, start small. Use it for the boring stuff. Keep your hands on the wheel for anything that matters.

I've been building workflows around AI tools for the past year — figuring out what actually sticks vs. what's just demo-worthy. If you're interested in that kind of practical approach, I share more on my profile.