dhool

@tldraw Genuine question, why should one use tldraw? Vs excalidraw

hey claude curl localhost:7236, then read ./github/workflows and create a tldraw explaining our release process

@davis7 People seem to forget swe is mostly about trade offs. Use the right tool for the job

The current "MCP is dead discourse" is my current favorite example of the really annoying "simplicity" brainrot that's been plaguing tech for a while On paper "just use a cli agents already know bash" sounds really good. It makes sense and is true, but it misses the bigger picture: > how do they know about the commands + their shape? > just put it in your agents md file or make a skill > ok, so then what happens if the cli's shape changes > well then u just update the markdown file > fine let's pretend people will actually do that (they will not) and this actually works, how are you going to scope authentication and authorization? > use the cli auth tools like what AWS has > ok so we need eng's or agents to be manually making sure every project switches the auth to the correct scopes for each project so that prod can't get vibe killed and then to connect to our internal services over cli in cloud agents we need to run basic agents in sandboxes now b/c we're not using mcp and... You get the point. It sounds really simple, and it is in the toy case, but in the real world an external API gated through MCP is actually way simpler, more secure, and manageable in a lot of cases. This post & article from @GergelyOrosz is a very good example: x.com/GergelyOrosz/s… It's the same thing with the $5 VPS or htmx or postgres or whatever other "simple" enlightened solution that works great in dumb indie hacker demos while making zero sense in the real world. I hate to break it to you guys, but there's no conspiracy. If modern tech solutions really were over complex slop do u really think these companies wouldn't take the free win to just do it the "simple" way? There are tons of problems with the tech, but it exists for a reason. I like CLIs a lot, and in a lot of cases skills make sense, clis make sense, etc. There are a lot of ways to do things, and they all solve different problems. But no, MCP is not dead. It or something like it isn't going anywhere. And there are a lot of good new ideas of how to make it better! Dumping an MCP with 60+ random tools into context sucks I completely agree and needs to be fixed. A lot of the code run solutions are very compelling like @RhysSullivan 's executor, cloudflare's code mode, and others. There are ways to fix this, but a "god mode bash tool" isn't it.

@asishcodes It’s good at frontend after the initial scaffolding and a good AGENTS.md is set. So I always start the frontend with Claude until basics are setup and then switch to codex

Codex seriously sucks at frontend. There’s something seriously wrong with ChatGPT 5.4 on the frontend part.

@WalterAtTheLab @powersync_ Can you explain with a real usecase?

@powersync_ Edge-native agents with local sync is underexplored — most agent stacks assume cloud connectivity, so the interesting failure modes happen offline. Are you seeing builders target autonomous IoT loops, or mostly mobile-first sync? 🧪

🦀 $500 BONUS PRIZE for the Best Rust SDK Submission in the PowerSync AI Hackathon! Our pre-alpha Rust SDK is ready for builders who want to push the boundaries of local-first and AI. Build embedded systems, IoT apps, or edge-native agents — all powered by real-time sync.

@devagrawal09 I am familiar with laminar in Scala, is it similar?

Solid 2.0 is gonna be the best framework for both humans and AI

@adxtyahq Are we allowed to use ai ?

just read a reddit post and a youtube breakdown of the Anthropic SWE interview loop never felt more dumb in my life. apparently you need to casually: - implement LRU cache with concurrency - build a web crawler - design LLM inference infra, KV cache management - reconstruct profiler traces from sampling profiler AI companies are interviewing like this now.

@MichaelArnaldi Effect full stack?

Today all hell broke loose, that’s great

@kitlangton @imjeehoo @motiondotdev @solid_js No effect ts?

@imjeehoo @motiondotdev Of course! @motiondotdev + @solid_js foreva

stupid sexy composer

@housecor the current generation of models you really never have to hand edit code, unless you have poor tooling and guard rails. Invest more time in the scaffolding of your projects if you are finding yourself tweaking the code

Claim: “I don’t need to understand code because AI generates it now.” Reality: Every abstraction leaks. When AI fails, tweaking the code is a superpower.

@jsedlacekjr It’s good after the basics are setup, but for 0->1 it’s really bad yea..

Codex is comically bad at user interface design

@devongovett You can still build without using AI, whos stopping you chief?

Is it just me or does this sound absolutely miserable to anyone else? Why would I want to spend my time writing requirements and reviewing code instead of directly building things? Building stuff is the fun part! How about we replace the boring parts like meetings and emails and status updates and issue triage. Engineers should be making tools to make *our* jobs better, not helping the idea guys replace us with slop.

@threepointone @cramforce Whats headline? , never understood the point , but also didn’t care enough to look deeper

ok @cramforce you cooked with just-bash. it's very good.

@wholyv taste is every thing in the ai era, and it ain’t lookig good for you chief

Okay I am impressed Kilocode > OpenCode If claude code can implement model agnosticism in their terminal then Claude Code > kilocode > opencode I don’t see any reason now why people would use Cursor or Antigravity

@ribelo_ @__morse It’s better than try catch

@__morse Is there a more fucked-up error-handling system than the one in Go? Is there anything people complain about more?

if Effect is Rust inside TypeScript errore.org is Go for TypeScript type safe error handling, the simple way

@IndraVahan A little late to the party mate, gpt-5.3-codex seems to have been nerfed ever since they started making it “faster” - opus 4.6 seems better since like a week ish now

almost fully on gpt-5.3-codex for coding now. openai really cooked. sorry opus but i’ve moved on

@addyosmani How do you automatically maintain them?

Tip: Be careful with /init. A good mental model is to treat AGENTS(.md) as a living list of codebase smells you haven't fixed yet rather than a permanent configuration. Auto-generated AGENTS(.md) files hurt agent performance and inflate costs because they duplicate what agents can already discover. Human-written files help only when they contain non-discoverable information - tooling gotchas, non-obvidous conventions, landmines. Every other line is noise. Beyond what to put in it, there's a structural problem worth naming: a single AGENTS(.md) at the root of your repo isn't sufficient for any codebase of real complexity. What you actually need is a hierarchy of AGENTS(.md) files - placed at the relevant directory or module level - automatically maintained so that each agent gets context scoped precisely to the code it's working in, rather than a monolithic file that conflates concerns across the entire project.

@VishwaTarak99 సూటి గా నిర్మొహమాటంగా చెప్పావ్ - కానీ నిజమే చెప్పావ్

Ila baaga uppu karam dengi add on ga oil fry cheskoni tintey eh mxxa protein radu . Nammakandi

@ryanflorence 🫡

@dharmesh 🫡

Have deleted my prior post about HubCode (the new project I'm tinkering with). Things I got wrong: 1) I didn't sufficiently acknowledge the popular open source (MIT license) opencode project, which I had used in order to get my proof of concept going. I did mention it, but not prominently enough. 2) I referenced the similarity of the HubCode terminal interface to Claude Code -- because that's the reference point that most people would understand (the project is meant for non-technical people). But, I understand there's some tension between Anthropic and Claude Code. Note: The project I've been working on has not launched in any way -- just something I'm tinkering with on my own local machine. The only reference to it are two social media posts. My apologies to the dev community and the opencode team. Turns out, I'm not going to be building a TUI anyways -- feedback from users has been that a web UI will be more helpful. But if I do build a TUI and do use opencode, rest-assured I will provide proper attribution and acknowledgment. Thanks.

@joesadoski @jayair 👍 still quite sad you can just copy with out attribution

@dhool_99 @jayair That license have been tested in court, it’s unclear if it would hold up.

2026, the year of the OpenCode forks