Leave a comment below to introduce yourself! You can talk about what brought you here, what you're learning, or just a fun fact about
yourself.Reply to someone's comment, either with a question or just a hello. 👋
Come back next week to greet our new members so you can one day earn our Warm Welcome Badge!
Top comments (510)
Welcome everyone to dev.to! Glad you are here and hope you are well! I am a DEV Community Moderator and my primary goal is to support everyone on dev.to and ensure everyone is having a good time!
To get started, I recommend reading this guide on making the most out of dev.to!
Get Started on Dev.to! A Beginner's Guide to Engage with the Community! 💡
Make sure to check out other resources here: dev.to/help/community-resources
Any questions about DEV and want to get answers from a DEV Community Mod? Leave a comment and come chat here!
Insights on Sloan and the MLH acquisition
Ask a DEV Community Mod! 🚀
Feel free to introduce yourself and welcome others by replying to at least 2 people! It would be greatly appreciated! :D
im harish, from a small village in india
Hello Harish, nice to meet you, I am Andrei and like you joined the dev.to Community just now 🎉
Hey Andrei. welcome to the dev.to community
welcome Harish
hello
Welcome, Harish! Great to have you here. What are you working on or hoping to learn? Plenty of friendly people around to point you at good resources.
welcome, harish
Welcome Harish!
nice to meet you
hola, yo soy de el salvador
Hi I'm Matt, a freelance web designer and developer from Cambridge, UK.
I'm pretty new to the industry, self-taught over the last couple of years. I've mostly been working in WordPress with a custom starter theme (gulp, scss, php, gsap) that I learned from an agency.
More recently, I've been experimenting with JAMstack and just rebuilt my portfolio site with it.
I'm really keen to keep learning more, get more exposure to real-life projects, and offer my support to small agencies who might need some extra help.
nice meeting you
Thanks Francis! Happy to be here — this looks like a great community to learn in 🙌
I’m Kael, from Singapore.
I mostly build tools for stock research and backtesting. Lately I’ve been working on
volas, a super-fast, Rust-backed, pandas-like DataFrame for OHLCV data, which has 254 built-in technical indicators. 😃I’m here to learn from people working with Python, Rust, data engineering, and market data tools. Nice to meet everyone.
Hey Kael, welcome to dev.to community
hello,Kael.Welcome to dev.
Hello Kael, Nice to meet you.
Hi all, I'm Scott from central Wisconsin, USA, and I just joined today. As you'll see if you look at my profile, I build WordPress websites for a living. Any of you who build websites, WordPress or not, knows that creating a good website requires many, many skills. So I thought, what better place to learn about a wide variety of topics, and that's how I ended up on dev.to.
Hi Scott, nice to meet you.
I've mostly been working in WordPress too. What tools do you use to build your WordPress websites?
Hey everyone! Senior sysadmin here. Just deployed my lightweight open-source behavioral network-socket monitor to GitHub. Excited to join the community and looking forward to sharing ideas and reviewing codebase architectures with you all!
Welcome! A lightweight behavioral socket monitor is a great niche to own, the kind of thing you end up reaching for constantly once it exists. What pushed you to build it from scratch rather than wire up existing tooling? Always good to see more people shipping open source here.
Amazing!
cool, how's going? I am starting with open source project as well, good to see more open source devs here.
Coolll
Hey everyone! I'm Amer, an IT and Network Infrastructure expert. I've been in the tech field since 1998, focusing on network maintenance, hardware, and recently, cybersecurity. I'm excited to join the DEV community, connect with fellow developers, and share insights from my long journey in tech. Looking forward to learning from you all! 🙂
Welcome @icoreux ! Dev is a great and welcoming community. Looking forward to reading about your journey!
welcome to dev.to
Hey everyone! I'm Jay, 24 years old from Munich, Germany.
I'm building APIs on the side — currently running a validation toolkit on RapidAPI (IBAN, email, phone, VAT, password breach-check and more). Also into trading and building my own AI dashboard.
Always happy to connect with other builders and developers. Looking forward to the community!
Hey Jonas. Welcome to the community. 😊
Wow, an API builder. Truly honored to meet you. What's your favorite API that you have build or used that have blown you away.
Awesome to have you here and have a great time here.😉
Thanks for the warm welcome! 😊
Honestly, the one that blew me away to build was the password breach-check endpoint — it uses HaveIBeenPwned's k-anonymity protocol, so only a 5-character SHA-1 hash prefix ever leaves your server. The full password never gets sent anywhere. Simple concept, but elegant privacy-by-design.
Currently working on a full validation toolkit (IBAN, VAT, email, phone, disposable email detection and more) — happy to share if you're curious!
Looking forward to connecting with everyone here 🙌
Wow, completely blown away by your sharings. You explained it like I'm 5. Sure, I would love to know more about it and all the best for it. Hope it comes and becomes Product of the Day on Github.
Hello Jonas :D
Hi 👋
Evgenii, from Valencia, Spain. I run a small web studio — React, TypeScript, Django, and lately a lot of LLM agents.
What brought me here: I keep finding the same problem on different clients' sites, and it isn't really a bug. People assume that if Googlebot can render their site, AI crawlers can too. They can't. I started writing about it, and this looked like the place where someone would actually argue with me. Please do.
Fun fact: I spent 12 years servicing medical lab analyzers before I touched the web. 300+ labs, coagulometers, immunoassay machines. Turns out debugging a React hydration issue and debugging a machine that hands someone the wrong blood test result use the same part of your brain. One of them just matters more.
Currently learning: what AI crawlers actually do — by reading server logs instead of vendor blog posts. Recommend it.
You asked for an argument, so: how long does the problem even last? Crawlers that can't render JS feels like a today-problem - the newer agent browsers drive real Chrome and render everything. My money is still on server-rendered HTML winning, just for a different reason: it's cheaper for the site than letting a thousand agents each run your JS. And 12 years of servicing lab analyzers is the best origin story in this thread.
Fair, and partly right. The agent browsers (ChatGPT agent, Perplexity's) do drive real Chrome and render everything. For those, JS is a non-issue.
But that's a narrow slice. The heavy lifting is GPTBot, ClaudeBot, Meta-ExternalAgent, and as of June 2026 none of the major AI crawlers render JS (Lantern's monitoring, plus a pile of May audits saying the same). The one exception is Gemini, and only because it rides Googlebot's infra. Meanwhile the volume is growing, not shrinking: GPTBot is up ~305% year over year per Cloudflare. So it's not a fading problem, it's scaling.
Which flips your framing, not mine: rendering at index scale is too expensive for them, so they don't, so the HTML has to carry the content. The day GPTBot renders JS at crawl scale I'll happily retract the post.
Your cost angle is the stronger version, honestly. "Cheaper for the site" beats "required for visibility" because it holds even in a fully-rendered world. Stealing that.
(And thanks re: the analyzers. Still not sure if it's a flex or a warning label.)
Deal - you take the cost angle, I'll take the correction. I was arguing from the session end (agent browsers); at crawl scale your numbers win. And 305% more crawling with zero rendering means they're scaling the cheap path, not working toward the expensive one. It's a flex, by the way.
"Scaling the cheap path, not working toward the expensive one" is the line I should've led with. 305% growth with zero rendering isn't a company that hasn't gotten to JS yet — it's a deliberate architecture. You don't scale something 3x while planning to replace it. Good arguing with you.
(Noted. Filing it under flex.)
Hi all! Joining the community as a Systems Engineer & Security Researcher. I focus on low-level tooling and hardened infrastructure. Currently hacking on PtyXL project. If you're into Rust, Tauri, CyberSecurity, or dev-tools, I’d love to connect and talk!
Hi. Welcome to the Community 😊.
Can you explain PtyXL project or is it Top Secret Mission that will be discussed later?😆
Awesome to see you! Enjoy the vibe. 🫡
Haha, absolutely - no corporate secrecy here, just a strict local-first philosophy. 😆
PtyXL is a modern local-first terminal client for SSH, SFTP, and Serial, written entirely in Rust with Tauri. Zero cloud sync, zero telemetry, and full data sovereignty for your infrastructure credentials.
We also just shipped the Provision Builder in v0.5.8, letting you orchestrate server configurations and pipelines by connecting logic tiles instead of writing raw shell scripts.
Appreciate the warm welcome! 🚀
This sounds too awesome. I would to try it and learn from you. Thank you so much for creating it. I wanna get my hands dirty in CLI or API building haha. Someday, I'll post about my projects too.
Awesome to have you. Keep building and breaking.😁
Thanks, that means a lot! Would love to have you try it out. If you hit any rough edges or have ideas, I'm all ears, still very early and shaping the roadmap based on exactly this kind of feedback. And yeah, definitely post about your projects - always good to see what people are building. 😄
👋 Hey DEV! Solo dev here — just started shipping small, single-purpose CLIs and figured I'd say hi.
Two so far, both MIT + on npm:
npx grokscope demo(no key needed to try it).Also just posted my first article here — what grokscope turned up digging into Bun vs Node. Excited to be around people who build things 👋
Hey Christo. Welcome to the community. 😊
These projects looks amazing. I would love to try them. Can you share lessons from your projects that you learned while building them.
Awesome to have you here and have a great time.😉
Thanks Konark, that means a lot 🙏
Biggest one: my tests were lying to me. Both had a full green test suite, but a couple of real bugs only turned up when I actually ran them against the live API — gemcatch was quietly sticking the prompt at the top of its own answers, and the free Gemini tier drops your result after 24h if nothing grabs it in time. I don't trust a green checkmark until I've done one real round-trip now.
Also learned to keep deps light the hard way — a native SQLite dependency blew up CI on one older Node + Windows combo, so when I added a cache to the other tool I kept it to a plain file with zero native deps.
One design thing too: make it runnable with no setup. grokscope is bring-your-own-key with a small per-query cost, so I added
npx grokscope demo— replays a real recorded run so you can see what it does before signing up for anything.Both are still pretty new so I'm finding rough edges as I go — if you try either, I'd genuinely love to hear what breaks!
Wow, amazing lessons. I would keep them in my mind before starting my projects. Thank you so much for sharing your lessons and thank you for sharing these projects. If I try I'll provide you an honest feedback.
Hey everyone! 👋
I'm Vishal, a software developer with 3+ years of experience and the solo founder of
VIGILANTE, a free GitHub-based accountability tool that helps developers build a consistent daily shipping habit.I'm here to learn, stay up to date with the latest technologies, and connect with fellow developers who enjoy building cool things.
I also love talking about software engineering, side projects, productivity, GitHub, and developer tooling. If you're working on something interesting or have questions about any of those topics, feel free to reach out. I'd be happy to help however I can.
And if building coding consistency is something you struggle with, give
VIGILANTEa try. I'd love to hear your feedback and learn how it can be improved for developers like you.Looking forward to learning and building together! 🚀
Hey Vishal. Welcome to the community. 😊
VIGILANTE feels amazing. I'm feeling like I'm secretly Batman. What was your biggest struggle while building VIGILANTE.
Awesome to have you here and have a great time here.😉
Hey Konark,
Thanks for your feedback on Vigilante, have you connected your account with it and monitored your consistency?
If yes, can you give an honest review here?
Biggest Struggle? - because of struggling in finishing a project, I started building Vigilante to be accountable of whatever I've planned for me.
Hey — Amin here.
I build open-source agent infra in Python (runtimes, local-first guards, graph RAG). Just published my first post about cutting repeat LLM calls in multi-agent graphs — happy to be here and learn from other builders.
Currently deep in native agent graphs vs the usual glue stacks. If you’re fighting token bills or prompt-injection at the edge, I’d love to compare notes.
Glad to meet you all 👋
Everyone posts agent architectures, nobody posts their invoice - so a first post about cutting repeat LLM calls is a good one. I went the opposite way: resend full history on every call by design (statelessness bought me control of the context) and just eat the cost. Curious what your approach gives up to save the calls.
Nice to meet you as well
That’s a clean trade — and I respect it.
If you own the full history on every call, you own the context window: no “what did the cache think was the same question,” no stale answer after the world moved, no mysterious skip. Statelessness buys auditability of what the model saw. Paying for that is a deliberate ops choice, not a bug.
What I give up to save the repeats is roughly the opposite of that guarantee:
I have to define “same enough.” Harmonic / semantic cache only helps when the intent is essentially a repeat. Unique turns, tool results that change the world, or “same words, new facts” should still climb. If the similarity gate is wrong, you either miss savings or (worse) reuse an answer that shouldn’t apply.
Staleness is my problem. Once something is cached, I own invalidation / TTL / scope — not the model. Full-history-every-time never has that class of bug.
Control moves from the prompt blob into the runtime. Ledger + cache keys instead of “the entire transcript is the source of truth.” That’s a win when graphs are multi-hop and the same honest question shows up again; it’s a loss if your product is “always re-ground on the full conversation.”
So I don’t treat cache as “never pay.” I treat it as: pay for novelty and for state changes; don’t pay twice for the same honest climb. Your design says: novelty detection is hard / risky, so always pay. Totally coherent — especially for agents where wrong reuse is worse than a fat invoice.
Curious where you draw the line inside the history itself (trim? summarize? always raw). That’s often where the bill actually hides even in the stateless path.
Happy to compare notes either way — invoice stories > architecture slides.
hello! I'm a mid-level software engineer. Came here just so that I can learn in public. Distilling down my knowledge of stuff I've worked on is a pretty great way to get new insights myself and I don't do it often enough. Lets see if I can keep it up on this account 😄
Hello @doubletime Learning in public here at Dev has been epic!
hello
Some comments may only be visible to logged-in visitors. Sign in to view all comments.