⚡ Building in Public

I used to think posting my work online didn’t matter. Now founders are reaching out saying they’d - “love to steal me from my current job.” Turns out consistency + authenticity really are the algorithm.

Looking for some feedback 👇 I've been building a small programming language called Rune to learn how languages work under the hood. Built from scratch with: - Lexer - Tokenizer - Recursive descent parser - AST generation - Tree-walk interpreter - Functions & closures It's definitely not production-ready 😅 Would love feedback on the syntax, architecture, docs.. or anything that looks odd. 🌐 rune.kapiljangid.pro 🐙 github.com/lang-rune

@TanzilaSha9574 Nothing special.. just a fun project rune.kapiljangid.pro

Hey builders 👋 What are you working on right now? Looking to connect with people building: 🚀 SaaS 🤖 AI Products ⚙️ Automation Tools 🌐 Web Apps 📱 Mobile Apps 💻 Developer Tools 🧠 Indie Projects Whether you're pre-launch, launched, or already getting users... Drop your project below 👇 Always interested in discovering new products and connecting with ambitious builders. 🚀 #buildinpublic #indiehackers #saas #ai #startups

@foxtomb232 Nothing special just a fun project rune.kapiljangid.pro

looking to connect people on X if you're into - building SaaS - vibe coding - AI tools - shipping in public - figuring it out as you go say Hi or drop what you're working on looking to follow active ones 👋

Not an actual AI service but I had built an personal CLI tool that does this part.. didn't actually get to finish it tho.. Core idea was fetching current trends in a given niche then generating 5-10 posts and then posting the selected post to X.. Similarly.. fetching posts from your timeline and then writing a reply to it...

Im looking for recs for an AI service that generates posts and replies for X. Got any good ones?

@achxvi Yo.. it's not a saas but working on a fun project with the goal of learning.. still in development tho rune.kapiljangid.pro

Drop your SaaS and I will give you feedback!

@uday_devops Tbh been juggling between different stack but I'd say loved what I built rune.kapiljangid.pro A language built because someone wanted to know how....

Which programming language made you fall in love with coding?

@foxtomb232 Ntg special tbh.. just been playing with code rune.kapiljangid.pro & getflowcms.com

What did you build this Wednesday? Drop your project URL 👇🏻 Let’s drive some traffic

@Cosmosfinite Yo.. sounds cool... Let's connect Here are a few things that I'm working on.. getflowcms.com rune.kapiljangid.pro

Hey! Founders and Builders! I really have to #connect with people in: 🍽️ Design 🚀 SaaS 📲 Automation 🧠 AI tools 📈 Ecom 📱 UI/UX 🔥 Web design/development 💻 Fullstack engineers Drop what you're building, who you are, your passion, builders grow by collaborating.👇

@ishivamgaur There are a lot bruh... Here are a few tho getflowcms.com rune.kapiljangid.pro

Be honest devs, how many side projects are still unfinished? 😭

@mshaheer2k3 Huh.. then why don't you try rune rune.kapiljangid.pro

@kjxcodez It ain't a programming language bruv 😭

which programming language is the worst?

Honestly, a lot of the work happens before I even start coding. I usually go through multiple revisions and iterations of the design, syntax and architecture first.. then start implementing. Once coding starts, it's mostly: Research > Prototype > Test > Refactor > Document 😅 Rune has already gone through several rewrites in old repos while I figured out what felt right.

@kjxcodez Side project discipline is different from work discipline. What does your actual working process look like on this?

A few months ago I got curious about how programming languages actually work. That curiosity somehow turned into Rune - a programming language I've been building from scratch. So far it has: - Custom lexer - Recursive descent parser - AST generation - Tree-walk interpreter - Closures & lexical scoping - REPL - VS Code extension - Documentation website Everything is open source. 🌐 Docs: rune.kapiljangid.pro 🐙 GitHub: github.com/lang-rune Now for the important part: Rune is still under active development. There are bugs. There are rough edges. There are probably design mistakes I haven't discovered yet. So if you enjoy breaking things: > Try Rune > Stress test it > Break it > Open issues > Share feedback > Suggest features > Contribute code I'd love to learn from people who know more about language design, compilers, interpreters, tooling, or just have interesting ideas. The goal isn't to build the next big language. The goal is to learn in public, build something fun, and understand how languages work under the hood. Docs → rune.kapiljangid.pro GitHub → github.com/lang-rune #buildinpublic #opensource #programming #python #compiler #interpreter #devlog #100DaysOfCode

Ehh damn...

@PratikSinhatwt Ntg productive atm.. but spent way too much time polishing tiny details nobody will notice anyway... here's my portfolio now: kapiljangid.pro curious what people think :)

2 am club? Anyone What are you building guys ??

spent way too much time polishing tiny details nobody will notice anyway... here's my portfolio now: kapiljangid.pro curious what people think :)

Feels like we're all racing to build agents. But I rarely see people talk about what happens after deployment. Question: What do people even use to understand why an agent did something? Logs? Tracing? Or mostly trial and error?

Something I've been wondering: Has anyone here had an AI agent do something in production that made you immediately think: "okay... we need guardrails" Not talking about benchmarks or local experiments. I mean real stuff: - wrong actions - repeated actions - weird tool calls - cost spikes - touching data it shouldn't - silent failures What happened? How did you figure out what the agent actually did?

Something I've been wondering: Has anyone here had an AI agent do something in production that made you immediately think: "okay... we need guardrails" Not talking about benchmarks or local experiments. I mean real stuff: - wrong actions - repeated actions - weird tool calls - cost spikes - touching data it shouldn't - silent failures What happened? How did you figure out what the agent actually did?