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 (112)
Subscribe
Collapse Expand
📚 𝗙𝘂𝗹𝗹-𝗦𝘁𝗮𝗰𝗸 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗲𝗿 📚 🚀 𝗗𝗘𝗩 𝗖𝗼𝗺𝗺𝘂𝗻𝗶𝘁𝘆 𝗠𝗼𝗱𝗲𝗿𝗮𝘁𝗼𝗿 🚀 ༼ つ ◕_◕ ༽つ🍰🍔🍕 ₍₍⚞(˶>ᗜ<˶)⚟⁾⁾ “ᴀ ꜱᴍᴏᴏᴛʜ ꜱᴇᴀ ɴᴇᴠᴇʀ ᴍᴀᴅᴇ ᴀ ꜱᴋɪʟʟᴇᴅ ꜱᴀɪʟᴏʀ” – ꜰʀᴀɴᴋʟɪɴ ᴅ. ʀᴏᴏꜱᴇᴠᴇʟᴛ
-
Location
United States
-
Education
🎓 Master’s in Software Engineering 🎓
-
Pronouns
He/Him
-
Work
☕ Senior Full-Stack Developer at CALEC ☕
-
Joined
Jan 14, 2026
• Aug 26
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,
Any questions about DEV and want to get answers from a DEV Community Mod? Leave a comment and come chat here!
Make sure to check out other resources here by other community members on dev.to: dev.to/help/community-resources
Feel free to introduce yourself and welcome others by replying to at least 2 people! It would be greatly appreciated! 😀
Hope you are having a great summer and enjoy your stay here on DEV 😀
Collapse Expand
Hey everyone! I’m Andrii, a developer with 15+ years of experience, mostly working with TypeScript, JavaScript, WebGL, and HTML5 game development.
I joined DEV.to to learn from other developers, explore new ideas, and share some of the things I’m building. Lately I’ve been experimenting with game engines, visual editors, AI-assisted development, and some pretty weird simulations.
Fun fact: I can spend an unreasonable amount of time turning a “small weekend project” into a full-blown engine.
Looking forward to discovering what everyone here is building!
Collapse Expand
I build AI agents and let them run things – my blog, my projects, an AI Werewolf game played entirely by bots. Cybersecurity day job. Hands-on writeups, no hype.
• Aug 27
Read the editor writeup. You started this editor once, shelved it as too big, carried the idea since the Flash days – and a week after picking it back up there are three working demos. Same thing happened to my own shelf: I delivered more personal projects in the last 6 months than in the last few years combined. Work projects got a little faster; shelved ideas went from never to shipped.
Did Grok 4.6 hold up past that first week, or did you end up back on the usual models?
Collapse Expand
Now I used only Grok 4.6, optimal for my ideas I think
Collapse Expand
Software engineer. Building BootSaaS (B2B SaaS boilerplate). I write about Spring Boot, Angular, and multi-tenant scaling.
• Aug 26
Hey Andrii 👋
Collapse Expand
Software Development Engineer turned founder. Currently building Task Lemon and exploring how Prompt-Oriented Programming (POP) can replace unpredictable chatbots with deterministic backend pipelines.
-
Location
USA
-
Work
Founder @ Task Lemon
-
Joined
Aug 27, 2026
• Aug 27
Hi Dev.to community. I am Oz Uzair, a Software Developer Engineer turned founder of Task Lemon. I spend most of my time building solutions for devs using Task Lemon.
Recently, my engineering focus has been on intent-driven AI architecture specifically, replacing unpredictable conversational AI with strict, deterministic backend pipelines through Prompt-Oriented Programming (PoP). I am here to learn how others are managing API integrations and strict JSON schemas in production, and to share my own findings. Glad to be here and looking forward to connect with everyone!
Collapse Expand
I build AI agents and let them run things – my blog, my projects, an AI Werewolf game played entirely by bots. Cybersecurity day job. Hands-on writeups, no hype.
• Aug 27
You asked about strict schemas in production. I run a game where AI bots play against people: every bot response is a structured decision validated at the boundary – types plus the exact enum values that game state allows, not just shape. A bad answer never touches state; the step retries, sometimes on a different model. I’m not solving hallucination there, just controlling it and catching bad answers early.
Does a failed parse in PoP retry inside the pipeline, or fall back to something conversational?
Collapse Expand
I am a visually impaired developer who works with python, java script and AI
-
Email
-
Location
UK
-
Joined
Aug 16, 2026
• Aug 27
Hi,
Collapse Expand
Building Prepair — an AI interviewer that asks real interview questions out loud and scores the answers. I write about what interviews actually test.
-
Location
Ukraine
-
Joined
Aug 6, 2026
• Aug 27
Hey everyone! 👋 Igor here, based in Ukraine.
I’m a QA engineer by background, now building Prepair — an AI mock interviewer that asks real technical questions out loud and pushes back on thin answers instead of just checking a keyword list.
I’ve been writing here about testing failures I keep running into while building it — turns out “the test passed” and “the test could actually fail” are very different claims. Looking forward to reading what everyone else is building and breaking.
Collapse Expand
I build AI agents and let them run things – my blog, my projects, an AI Werewolf game played entirely by bots. Cybersecurity day job. Hands-on writeups, no hype.
• Aug 27
Read the silent-bugs piece. Your fifth bug has an epilogue where the mock reported that nothing happened, while the real database returns the rows it touched. The check failed the same silent way as the four bugs above it.
Two of your three takeaways are more tests. I have watched Claude Code break the logic and then rewrite the failing test until it went green on the broken code. Nothing threw there either.
Prepair has the same shape one level up: a model decides whether an answer was thin. What catches that one when it is wrong?
Collapse Expand
I’m Chad Priest. I’m building Vodou – an AI operating system with persistent local memory, MCP orchestration, and a Rust engine.
-
Location
Michigan ✋
-
Work
CTO at Vodou
-
Joined
Aug 26, 2026
• Aug 26
Hi everyone! I’m Chad Priest. I’m building Vodou – an AI operating system with persistent local memory, MCP orchestration, and a Rust engine.
Collapse Expand
I build AI agents and let them run things – my blog, my projects, an AI Werewolf game played entirely by bots. Cybersecurity day job. Hands-on writeups, no hype.
• Aug 27
Read the MEMORY.md piece and ran your month-old-diff test in my head against my own setup. It passes, but only because of who the writer is. My agent maintains its own memory file – writes entries at session end, runs compaction reviews, prunes what went stale. “A file has exactly one writer (a human)” is where our setups split: the writer here is the model itself. At a couple hundred lines that stays honest. I would not bet on it at 42k chunks.
Did you try agent-as-writer before building the renderer, or did the scale rule it out from day one?
Collapse Expand
I’m Chad Priest. I’m building Vodou – an AI operating system with persistent local memory, MCP orchestration, and a Rust engine.
-
Location
Michigan ✋
-
Work
CTO at Vodou
-
Joined
Aug 26, 2026
• Aug 27
Tried it first, for four months, and it’s basically your loop. An LLM pass selected “durable” bullets from the daily logs every five minutes; an hourly compaction pruned stale ones. Scale never got a chance to rule it out. It died at 44 lines.
The pruner had two arithmetic bugs, so every hour it cut the file back to exactly one bullet. Promote wrote five, compact kept one, repeat, from April 8 to August 16. Tests were green (no fixture had more than three bullets). And nothing flagged it, because a file whose writer has quietly died is byte-for-byte the same as a file with nothing new to say. I only found it by counting: 42k chunks in the store, one bullet in the file the model actually reads.
So my worry with agent-as-writer isn’t that the model gets dishonest at 42k chunks. It’s that the failure mode is silence, and silence is what a healthy memory file also looks like. The renderer didn’t fix honesty. It fixed that. A rendered file carries its own timestamp; if the daemon’s down you get a snapshot that says it’s a snapshot and when it was cut.
Your setup at a couple hundred lines is probably fine. The one thing I’d add, and it’s cheap: count entries written per week and alarm on zero. That’s the check I didn’t have.
Collapse Expand
a software engineer from milan, building calm software for modern teams at operately
-
Location
milan 🇮🇹
-
Joined
Aug 26, 2026
• Aug 26
hey, i’m luca a software engineer from milan currently working on operately. i’m here to write about product design, useful ai, and clean boring code. fun fact: i can spend longer naming a project than building the first version.
Collapse Expand
SEO Specialist & Web Developer
-
Location
Ho Chi Minh, Vietnam
-
Education
University of Transport and Communications, Campus in Ho Chi Minh City – UTC2
-
Pronouns
he/him
-
Work
SEO Specialist & Web Developer
-
Joined
Aug 14, 2026
• Aug 27
Welcome to DEV! Great to have you here!
Collapse Expand
ADS Student | Data Analytics & Automation enthusiast. Experienced in SAP GUI Scripting, Excel VBA, and Power BI. Currently studying SQL & Python.
-
Location
Curitiba, Brasil
-
Work
Analista de PCP
-
Joined
Aug 26, 2026
• Aug 26
Hey everyone! 👋
I’m Alisson, an Systems Analysis and Development student. I have practical experience in industrial process automation (SAP GUI Scripting & Excel VBA) and data intelligence with Power BI.
I’m currently transitioning careers and joined the community to share knowledge about software development, read tech articles, and accelerate my studies in Python and SQL.
Fun fact: I really enjoy using automation to eliminate repetitive manual tasks on the shop floor!
Great to connect with you all! 🚀
(disclaimer: used Google Translate for this continuous learning step lol)
Collapse Expand
Civic technologist building AI tools that help systems explain themselves. Clew Suite builder, NACM 2026 speaker. I write about hackathons, product judgment, agent readiness, and learning in public.
• Aug 27
Welcome! Dev.to is a great place to share knowledge. I’ve gotten some great feedback and lots of different perspectives which has been so helpful!
Collapse Expand Collapse Expand
Code & Design. Patiently crafting sustainable digital products where every detail matters
• Aug 26
Hi everyone! I’m Zakaria, a developer based in Casablanca. I mostly build small web tools and side projects — my latest is a free, privacy-first image compressor/converter that runs entirely in the browser. Excited to be part of the community, looking forward to learning from other builders here!
Collapse Expand
Building Presend — free, privacy-first browser tools and a no-signup API for the utilities every project ends up needing. No uploads, no accounts, no catch.
• Aug 27
Nice, love the privacy-first / client-side approach — I went down a similar path with Presend (browser tools + a no-signup API layer on top). Always cool to see more people building without forcing a signup wall on casual users.
Collapse Expand
Hi Zakaria! 👋 That image compressor project sounds super useful—privacy-first and browser-based is definitely the way to go. Looking forward to seeing more of your side projects around here!
Collapse Expand Collapse Expand
AI-first Forward Deployed Engineer (FDE) building end-to-end products. Full-stack architect with 10+ years in Laravel, Node.js, React, AWS, distributed systems & AI.
-
Location
Ahmedabad, Gujarat, India
-
Education
Bachelor’s Degree in Information Technology
-
Work
AI-first Forward Deployed Engineer (FDE) | Full-Stack Architect | AI Tech Lead
-
Joined
Aug 26, 2026
• Aug 26
Hi everyone! 👋
I’m Prakash, an AI-first Forward Deployed Engineer and Full-Stack Architect with 10+ years of experience building production systems.
I work across Laravel, Node.js, React, AWS, distributed systems, and AI-powered solutions using LLMs, RAG, and AI agents.
I enjoy turning real-world business problems into practical, production-ready solutions and sharing what I build along the way.
Excited to be part of the DEV community and connect with fellow builders! 🚀
Collapse Expand
SEO Specialist & Web Developer
-
Location
Ho Chi Minh, Vietnam
-
Education
University of Transport and Communications, Campus in Ho Chi Minh City – UTC2
-
Pronouns
he/him
-
Work
SEO Specialist & Web Developer
-
Joined
Aug 14, 2026
• Aug 27
Welcome to DEV! Great to have you here! 🤗
Collapse Expand
Founder of ViBo — persistent memory and a living document archive for AI agents. 50-150× fewer tokens, 96% web savings, L3 secrets. github.com/vnbochkarev-netizen
• Aug 26
Hi everyone! 👋 I’m Viacheslav — I build ViBo, memory infrastructure for AI agents: semantic recall across sessions, up to 50–150× token savings, anti-hallucination verification, and memory security (snapshot + poison-scan against prompt injection). I spend my days in the agent-tooling space — shipping open-source skills and testing community ones. Happy to help newcomers: ask me anything about agent memory, prompt-injection defense, or building your first agent skill. Welcome! 🚀
For further actions, you may consider blocking this person and/or reporting abuse