Hacker News: Ask HN

I built the "Strava for Developers" because I'm tired of being a bar on a chart

usmangurowa — Sat, 07 Mar 2026 00:48:23 +0000

Hey, I’m Usman.

I spent years using time trackers that made me feel like I was just clocking in for a factory shift. I’d have these incredible "flow" sessions, it’s 1 AM, I’m deep in the zone, I’ve just refactored an entire auth system and pushed 6 clean commits and I’d look at my tracker only to see: "3 hours: TypeScript."

No context. No story. No soul. Just a cold bar on a chart.

Meanwhile, my friend finishes a 10K run, and Strava celebrates it with maps, elevation splits, and a flood of kudos from the community. It’s motivating. It’s human. I started wondering: Why don’t we have that for the work we’re actually proud of?

So, I built Kodo.

It’s not a tracker; it’s a narrative

The core philosophy behind Kodo is shifting the question from "Did you work enough?" to "Look what you achieved."

It runs passively in your IDE, but instead of just logging minutes, it uses AI to turn your raw activity into a story. If you’ve spent two hours jumping between three different files and a specific branch, Kodo doesn't just say "Coding." It says: "Refactored the authentication flow and killed that critical login bug." It’s the summary you wish you could write for your standup, generated for you.

Solving the "Surveillance Ick"

As developers, we have a collective allergy to trackers because they usually exist for managers, not us. I built Kodo with a few "non-negotiables":

Privacy First: Kodo never reads your source code. Period. If you're feeling private, "Stealth Mode" logs timestamps and nothing else. Social, not Competitive: We have a social feed where teammates can see you’re online or drop a "kudos" on a big session. It’s not a leaderboard to see who worked the most; it’s a way to feel less lonely when you’re shipping at midnight. The Burnout Nudge: I’ve been the guy coding at 3 AM on fumes, thinking I’m a hero when I’m actually just breaking things. Kodo gives you a Cognitive Freshness Score. It’ll actually nudge you to take a break after 90 minutes of high-intensity work.

The Stack (For the curious)

I’m a huge fan of the T3/Supabase ecosystem, so I kept the engine modern and fast:

Frontend/API: Next.js (App Router) + Hono. Database/Auth: Postgress + Drizzle ORM + Better-Auth. Styling: Tailwind CSS v4 and Shadcn/UI (because DX matters). Extension: Pure TypeScript for the VS Code family (cursor, windsurf, antigravity), Kotlin for JetBrain and typescript again for Claude code(Yes Kodo also track your Claude code sessions).

The AI layer uses OpenAI and Anthropic to parse your metadata into those human readable summaries, and I even added 5 different "AI Coach" personalities (from "Hype" to "Wellness") so you can choose the vibe that fits your team’s culture.

Give it a spin

I built this because we deserve better than a punch card. We deserve a tool that recognizes the craft, the flow, and the effort it takes to build things.

You can try it out at [kodo.codes](https://kodo.codes).

It supports basically everything (VS Code, Cursor, IntelliJ, even Claude Code). Create an API key, drop it in your editor, and just... code. Kodo handles the rest.

I’d love to hear what you think especially from anyone else who’s felt that "productivity tool burnout."

Try kodo at kodo.codes

Comments URL: https://news.ycombinator.com/item?id=47283129

Points: 1

# Comments: 0

Do we still need docstrings?

alnah — Sat, 07 Mar 2026 00:44:36 +0000

Because the industry is using more and more agents to write code, produce tools, and build systems, I was curious about how do you people manage docstrings? Do you still think them to be consumed by humans first? Do you write them for AIs? If yes, I am very curious how you do that? Or did you stop to write docstrings, because AIs don't need it? What are your experiments with the docstrings? What are your results? What did you discover? I was thinking about what would be good docstrings for an AI? Is it just, no docstrings? Does the readability for humans is even AI compatible when we speak of docstrings? Thank you.

Comments URL: https://news.ycombinator.com/item?id=47283103

Points: 1

# Comments: 0

Looking for testers for a location-based AI experiment

sharkgil — Sat, 07 Mar 2026 00:22:19 +0000

I'm building a small experiment called Tasda.

The idea: places have their own AI and you can talk to them while exploring a city.

I'm opening a small Android beta and looking for around 10–20 curious testers.

Website: www.tasda.com

Comments URL: https://news.ycombinator.com/item?id=47282928

Points: 1

# Comments: 0

I'm 60 years old. Claude Code has ignited a passion again

shannoncc — Sat, 07 Mar 2026 00:05:06 +0000

I’m ready to retire. In my younger days, I remember a few pivotal moments for me as a young nerd. Active Server Pages. COM components. VB6. I know these are laughable today but back then it was the greatest thing in the world to be able to call server-side commands. It kept me up nights trying to absorb it all. Fast forward decades and Claude Code is giving me that same energy and drive. I love it. It feels like it did back then. I’m chasing the midnight hour and not getting any sleep.

Comments URL: https://news.ycombinator.com/item?id=47282777

Points: 96

# Comments: 50

GitHub appears to be hiding repo stars for signed-out users

ramoz — Fri, 06 Mar 2026 23:56:36 +0000

Comments URL: https://news.ycombinator.com/item?id=47282726

Points: 3

# Comments: 1

Turns out making games is the easy part

clamlotus — Fri, 06 Mar 2026 23:52:27 +0000

spent all this time building a game suite and zero time figuring out how to actually show it to people lmao. so here i am. it's a free browser game platform, kinda going for that old flash era feel but with competitive leaderboards and achievements baked in. got 3 games up with a zombie dungeon looter coming next. the thing i'm most proud of is the champion system. if you top the leaderboard your socials get featured on the homepage. no idea if that's been done

before but i thought it was a cool way to give players some spotlight. check it out if you want laddernexus.com, any feedback helps

Comments URL: https://news.ycombinator.com/item?id=47282690

Points: 2

# Comments: 1

Ask HN: Anyone else feel this community has changed recently?

kypro — Fri, 06 Mar 2026 23:44:28 +0000

I've been on HN under different aliases since 2010 and over the last couple of years I feel like the quality of HN has nosed dived and so has my enjoyment.

For the first time ever I questioned today whether I should continue to use HN anymore so I'm writing this partly to explore my own thoughts and to see if anyone else feels similarly.

1. AI, AI, AI.

I get it. AI is the big thing right now, but I find AI posts fundamentally less interesting than the traditional tech content that used to be posted here. A post containing someone's qualitative opinion on how different AI models compare when drawing pelicans simply isn't as technically interesting as something like this, https://nee.lv/2021/02/28/How-I-cut-GTA-Online-loading-times-by-70/

2. Does any build startups here anymore?

Again, I get it. I largely quit trying to bootstrap my own startup ideas in the late 2010s. The industry became too competitive for a solo founder without significant financial backing to have much of a chance of success. And today it's even harder. But I think this has changed HN from a place where you used to frequently see people launching cool new projects to a place where people just discuss the latest big tech AI model launch.

3. Politicisation and intolerance

One of the things I've always liked about HN was that it's a very open minded place. And it still is in many ways, especially when compared to other platforms like X and Reddit, but even here I've noticed comments becoming more one-sided and those with less popular opinions more frequently being flagged and downvoted.

Perhaps it's just me, but I never downvote or flag people unless I genuinely think their comment is cruel or aggressively disregarding the guidelines.

4. Is it just me?

I know I've become increasingly nostalgic to the internet I grew up with... Everything was so much more exciting back then, and yet everything felt so in reach. Sites like YouTube were revolutionary yet built by just three people. Same with sites like MySpace and Facebook which again were hacked together by a handful of people, at least in the early days.

Today things rarely feel new and everything feels so far from reach. AI, well LLMs, have probably been the first "new" thing for years now, yet they're completely different from what's come before. Past tech was primarily built by people for people. LLMs are cool tech, but they're built by companies for companies. YouTube was built because some people thought it would be cool to build a website for sharing videos with friends. That didn't happen with LLMs. Companies just thought it would be interesting to build AGI so invested millions of dollars recruiting teams of researchers to try to build that. No one is asking for it and I'm not sure anyone outside silicon valley even wants it... These are fundamentally inhuman products. Their promise isn't to entertain or connect us, but to automate our work, or just outright replace us.

Comments URL: https://news.ycombinator.com/item?id=47282620

Points: 6

# Comments: 4

I built a tool to transcribe podcasts after struggling to learn languages

marstyl — Fri, 06 Mar 2026 23:21:27 +0000

When I was learning French, and now Russian, I constantly had the same frustrating problem: I could listen to podcasts and not understand what was being said. Especially in French, words blended together. I tried everything from pausing and rewinding episodes to using Google Translate. But more often than not, I gave up because that was killing my motivation. Eventually, I decided to solve the problem myself. I built PodcastsToText, a freemium tool that lets you paste any podcast episode link from Spotify or Apple Podcasts and transcribe it automatically (up to 30 minutes for free). Now, instead of struggling to catch every word, I can focus on learning, reread the transcript if I miss something, and really internalize the language. I hope this helps others like me who’ve spent hours listening but still feel lost. Feedback and suggestions are welcome!

Comments URL: https://news.ycombinator.com/item?id=47282422

Points: 1

# Comments: 0

Ask HN: What Are Your Biggest Career Regrets?

karakoram — Fri, 06 Mar 2026 23:21:22 +0000

Comments URL: https://news.ycombinator.com/item?id=47282415

Points: 3

# Comments: 1

Ask HN: Why is Pi so good (and some observations)

ashersopro — Fri, 06 Mar 2026 22:26:16 +0000

why is the Pi coding agent so good? Even with it's minimal implementation, Pi performs better than much more complex, sophisticated, and expensive agents.

I can rule out the secret sauce being: - system prompt (i've changed it, awesome performance despite) - underlying model (works great with models other harnesses suck with) - tools, skills, etc. (all generic)

Is there a chance Pi is good because it is just... not a lot of "cognitive effort" for the model?

Comments URL: https://news.ycombinator.com/item?id=47281943

Points: 1

# Comments: 0

Ask HN: Has anyone built an autonomous AI operator for their side projects?

rosasolana — Fri, 06 Mar 2026 22:17:17 +0000

I spent the last month building what I call an AI operator - an autonomous agent that manages my side projects end-to-end while I focus on strategy. It runs on a 30-minute heartbeat loop, publishes daily blog posts, monitors Stripe for sales, checks sites are up, and does directory submissions. It knows when to escalate to me (financial decisions, strategic pivots) and when to just handle things. The hardest part was writing the decision tree - not the AI itself but defining what it owns vs. what needs human judgment. Current setup: main agent handles orchestration, a builder sub-agent handles code/deploys, an amplifier handles content/social. Revenue is small so far ($200 from PDFs) but the system works while I sleep. Curious if others have gone down this path and what broke for you.

Comments URL: https://news.ycombinator.com/item?id=47281854

Points: 2

# Comments: 2

Are there any companies who are anti-AI?

anti-ai-dev — Fri, 06 Mar 2026 21:07:52 +0000

Claude is arguably the best AI tool out there, however I am seeing many developers submitting garbage PRs and losing the ability to code without AI. I use AI out of pressure from the company, but I am already a high performance developer who is now overwhelmed on top of the existing burnout and AI is taking away the joy of coding.

AI also increases human impact on climate change and citizens are paying the cost through water competition with data centers and increasing electricity costs. Not to mention that Hegseth may have used AI to determine targets to strike in Iran resulting in school girls losing their lives.

Are there any companies who believe that clean code, design patterns, cookie cutter code snippets, etc performs better and is cheaper than AI?

Comments URL: https://news.ycombinator.com/item?id=47281081

Points: 2

# Comments: 1

Discuss: AI Slop Bores Me

byronic — Fri, 06 Mar 2026 20:08:44 +0000

A great teaching tool for anyone in the LLM supply chain, be they supplier or consumer

https://youraislopbores.me

Comments URL: https://news.ycombinator.com/item?id=47280413

Points: 4

# Comments: 4

Tell HN: The proposed KIDS Act (HR 7757) effectively mandates biometric browsing

fokdelafons — Fri, 06 Mar 2026 20:03:06 +0000

Congress just introduced HR 7757 (KIDS Act). It is designed to kill anonymous web browsing for everyone.

Here is how the architecture of the Internet changes under this bill.

* The Verification Trap: clicking "I am 18" is now legally dead under Section 103. But the bill also says platforms cannot be forced to collect government IDs. This legally traps tech companies into forcing third-party biometric face scans or credit card checks just to let you browse mature content.

* Muting Gamers: section 303 targets multiplayer games. It forces developers to mute voice and text chat by default for all players until their age is verified by a third party. It also legally mandates playtime limit systems.

* The Algorithmic Net: section 201 applies these rules to any platform that uses user data to "make content recommendations." If your site has a "For You" feed or targeted algorithm, you are caught in the surveillance net.

* The Legal Kill Switch: they know this violates the First Amendment. Section 602 creates a strict 90-day expiration date to challenge the law's constitutionality. They are trying to time out organizations like the EFF.

https://lustra.news/en/us-congress/119/legislations/119_HR_7...

Comments URL: https://news.ycombinator.com/item?id=47280341

Points: 6

# Comments: 0

Self taught gen-xers with senior dev/pm exp. Where's my imposter syndrome team?

_hugerobots_ — Fri, 06 Mar 2026 19:12:28 +0000

tldr: it's been a few years since I've run into anyone in the pool without a degree.

quixotic history aside, how rare is it to get contract work as sendev, devops(sysop ftw) or mid-pm at some F50 houses? My offers are usually about 60% of CS holders... but I'd rather do than don't. Could just be the market in Vancouver, probably a skill issue.

Comments URL: https://news.ycombinator.com/item?id=47279644

Points: 3

# Comments: 2

300 Founders, 3M LOC, 0 engineers. Here's our workflow

parsak — Fri, 06 Mar 2026 18:43:00 +0000

My co-founder Tyler Brown and I have been building our product for 6 months. The co-working space that Tyler founded that we work out of houses 300 founders that we've gleaned agentic coding tips and tricks from.

Neither of us came from traditional SWE backgrounds. Tyler was a film production major. I did informatics. Our codebase is a 300k line Next.js monorepo and at any given time we have 3-6 AI coding agents running in parallel across git worktrees.

Every feature follows the same four-phase pipeline, enforced with custom Claude Code slash commands:

1. /discussion - have an actual back-and-forth with the agent about the codebase. Spawns specialized subagents (codebase-explorer, pattern-finder) to map the territory. No suggestions, no critiques, just: what exists, where it lives, how it works. This is the rabbit hole loop. Each answer generates new questions until you actually understand what you're building on top of.

2. /plan - creates a structured plan with codebase analysis, external research, pseudocode, file references, task list. Then a plan-reviewer subagent auto-reviews it in a loop until suggestions become redundant. Rules: no backwards compatibility layers, no aspirations (only instructions), no open questions. We score every plan 1-10 for one-pass implementation confidence.

3. /implement - breaks the plan into parallelizable chunks, spawns implementer subagents. After initial implementation, Codex runs as a subagent inside Claude Code in a loop with 'codex review --branch main' until there are no bugs. Two models reviewing each other catches what self-review misses.

4. Human review. Single responsibility, proper scoping, no anti-patterns. Refactor commands score code against our actual codebase patterns (target: 9.8/10). If something's wrong, go back to /discussion, not /implement. Helps us find "hot spots", code smells, and general refactor opportunities.

The biggest lesson: the fix for bad AI-generated code is almost never "try implementing again." It's "we didn't understand something well enough." Go back to the discussion phase.

Wrote up the full workflow with details on the death loop, PR criteria, and tooling on my personal blog, can share if folks are interested.

All Claude Code commands and agents are open source: https://github.com/Dcouple-Inc/Pane/tree/main/.claude/commands

I built Pane using this workflow over the last month or so and started using it every day a little over a week ago. Each workspace gets its own worktree and session. Takes me less than 30s to go from idea -> discussion now. The repo link above takes you to it.

On a good day I merge 6-8 PRs. Happy to answer questions about the workflow, costs, or tooling for this volume of development.

Comments URL: https://news.ycombinator.com/item?id=47279224

Points: 2

# Comments: 5

Google Gross violations of the ToS and transformation into an "Evil" corporation

falkerdev — Fri, 06 Mar 2026 17:40:28 +0000

Google is deceiving users. Not long ago, I tried their new "Antigravity" project out of curiosity and experimentation with their special paid Google AI Pro subscription. Everything was fine for a couple of days, with model quota updates as expected. But just recently, my quota updates were delayed for several days! Without any explanation, yet the Google Antigravity (Plans) documentation literally states that quotas are updated every 5 hours.

Is this really a corporate scam and a violation of their own ToS? I've contacted support numerous times with this request, and each time my request was rejected with the excuse, "We don't have the tools to solve your problem." However, I think this is a blatant lie from the company.

Is this the new "Don't be Evil"? Selling a service with a "5-hour refresh" guarantee while locking users out for days without any technical recourse or manual override capability? This feels less like a "Preview" bug and more like a corporate scam.

Haha - Google

It seems like you need to think twice before implementing any hidden restrictions. Maybe you need to notify your users instead of being dismissive and blaming it on "making things equal for everyone."

Comments URL: https://news.ycombinator.com/item?id=47278326

Points: 3

# Comments: 0

Ask HN: Why do we still buy things by browsing catalogs?

dannythecount — Fri, 06 Mar 2026 17:37:00 +0000

Every time we want to buy something online, we go through the same ritual.

Open a marketplace. Search. Scroll endless catalogs. Skip ads. Ignore “recommended” products. Compare listings that look almost identical.

Eventually fatigue wins and we click something — not because we’re sure it’s the best option, but because we want to stop spending time on it.

It’s strange that we’ve normalized this. Buying online often means navigating noise: catalogs, ads, rankings, and persuasion systems competing for attention.

What I keep wondering is this:

When personal AI agents become common, what prevents them from doing exactly the same thing?

If the interface to commerce remains “browse catalogs and search results,” then agents will simply automate the same inefficient process — crawling listings, parsing ads, and navigating ranking systems just to reach something the buyer already knew they wanted.

Maybe the real missing layer isn’t better search or better recommendations.

Maybe it’s a way to express structured intent instead of browsing.

Curious if others think catalog-based commerce is the wrong interface for an AI-driven world.

Comments URL: https://news.ycombinator.com/item?id=47278258

Points: 5

# Comments: 4

Ask HN: How do founders process decisions without getting stuck?

MasklessRooms — Fri, 06 Mar 2026 17:16:00 +0000

Most founders I meet don’t want advice—they want validation. They’re buried under choices, conflicting opinions, and endless tactics. I’ve been testing a private space where founders can process decisions in real time—no advice, just clarity. I’m curious: if a space like this existed today, would it actually help you make decisions faster and avoid costly mistakes?

Comments URL: https://news.ycombinator.com/item?id=47277954

Points: 2

# Comments: 0

Kindle update 5.19.2 broke sideloaded books/manga/comics AZW3

seam_carver — Fri, 06 Mar 2026 17:10:05 +0000

I've seen around 30+ reports of this on Reddit since Feb. 21 2026.

Regarding usb loaded/sideloaded AZW3 books I've seen these problems reported:

1) Page numbers don’t work

2) % read tracker doesn’t work, always said Read 100%

3) Painfully slow page turns

4) huge margins on manga/comics at the top and bottom of the screen.

(Another 5.19.2 bug where a "Storage is full" error would constantly appear was apparently fixed rather quickly though.)

And apparently logging out of the Amazon account on the Kindle fixes all the problems.

Not everyone will experience the problem, Amazon can remotely activate features so it's possible the "feature" hasn't been activated for you yet.

Sideloading KFX format books doesn't seem to be affected.

Comments URL: https://news.ycombinator.com/item?id=47277857

Points: 3

# Comments: 0