HackerNews

Qwen 3.8 Max Preview

Article URL: https://www.qwencloud.com/pricing/token-plan Comments URL: https://news.ycombinator.com/item?id=48966190 Points: 84 # Comments: 35
HackerNews

Claude Code uses Bun written in Rust now

Article URL: https://simonwillison.net/2026/Jul/19/claude-code-in-bun-in-rust/ Comments URL: https://news.ycombinator.com/item?id=48966569 Points: 98 # Comments: 100
HackerNews

Qwen 3.8 Max

Article URL: https://qwen.ai/home Comments URL: https://news.ycombinator.com/item?id=48966754 Points: 23 # Comments: 1
HackerNews

The death and rebirth of my home server

Article URL: https://sgt.hootr.club/blog/home-server-rebirth/ Comments URL: https://news.ycombinator.com/item?id=48966769 Points: 5 # Comments: 0
🌍 Dev.to

Stop navigating dashboards: Automating Douyin Local Life with MCP

I spent a decade watching developers build incredibly complex integrations for massive ecosystems, only to watch them break because someone changed an undocumented field in a JSON response. Whether it's the Amazon API or China's Douyin ecosystem, the problem is always the same: you have powerful data, but your interface—the Merchant Center—is designed for humans clicking buttons, not for automated logic. When I saw how messy managing local commerce operations can be—handling split-second coup...
🌍 Dev.to

Cloudflare wasn't blocking our IP. It was blocking our browser.

A 20-minute experiment that saved us from buying residential proxies, and the escalation ladder we found underneath. At IntelDif, we crawl competitor pages on a schedule, diff them, and alert our users only when something meaningful changes. That business dies if the crawler can't see the page. So, when one of the domains we monitor started returning Cloudflare's "Attention Required" challenge on every crawl, it immediately went to the top of the queue. The internet has one answer fo...
🌍 Dev.to

OpenAI-compatible API first-call smoke test before you scale a workflow

When a team adds a new AI API route, the first milestone should be small. Do not start with a production migration, a large benchmark, or a full agent rollout. Start with one successful request that proves the base URL, API key, model name, balance, request shape, and logs are all working. Disclosure: I work with ModelRouter. ModelRouter is an independent third-party OpenAI-compatible AI API gateway. It is not an official service from OpenAI, Anthropic, Google, DeepSeek, or any model provi...
🌍 Dev.to

n8n Salesforce Node: Automate CRM Records, Leads, and Opportunities (Free Workflow JSON)

Salesforce is the world's most-used CRM, and if you're running workflows that touch leads, contacts, accounts, or opportunities, the n8n Salesforce node can automate the tedious parts — data entry, record syncing, pipeline updates, and alert routing — without a custom API integration or an expensive middleware tool. This guide covers everything you need to know to use the Salesforce node in n8n: what it can do, how to authenticate, common patterns, and the gotchas that will trip you up. ...
🌍 Dev.to

I was hired before I could speak German, and that CV would be filtered out today

My German was good enough to order bread. It was not good enough to defend an idea. There is a room in Germany I still think about, an interview I had no business sitting in. No local network, and an accent that made every sentence take longer than it needed to. Everything on paper said no. Somebody in that room said yes anyway. Sixteen years of building software have happened since. Four ambassador titles, earned honestly. None of it exists if one person had not looked past a fi...
🌍 Dev.to

I Built a Deep Learning Framework from Scratch in C++ and CUDA (And Beat PyTorch's Speed Multiple Run)

If you work in AI today, you are almost certainly using PyTorch or TensorFlow. They are incredible tools, but they are also massive black boxes. As an AI developer, I realized that relying solely on these frameworks meant I didn't truly understand the underlying hardware realities, memory mechanics, or how the math actually maps to GPU acceleration. So, I decided to strip away the abstractions. Over the last few months, I built Aakaar —a deep learning framework developed completely from...
🌍 Dev.to

Building and Deploying a Role-Based Django PWA: From Starter Template to Production

I recently built a full volunteer-opportunity board on top of a Django starter template — complete with role-based accounts, image uploads, a Progressive Web App layer, and a real production deployment. This post walks through what I built, the decisions behind it, and — more usefully — the deployment problems I actually hit and how I solved them, since that's the part tutorials usually skip. Live app: https://volunteer-board-nccd.onrender.com Source code: https://github.com/13getid/d...
🌍 Dev.to

PyGo: A Deep Learning Framework Where Go Calls Python Calls C++

[Project] PyGo – embedding CPython inside a Go process to build a deep learning framework I've been working on something a bit unusual: a deep learning framework where Go is the top-level API, Python handles autograd and the model zoo, and C++/CUDA does the raw compute. The architecture looks like this: Go API → CGo bridge → CPython (embedded) → pybind11 → CUDA/AVX-512 kernels The key insight: instead of a Python sidecar in every pod, CPython runs inside the Go binary....
🌍 Dev.to

Your n8n Workflow Passed Schema Validation—and Updated the Wrong Customer

A workflow can finish successfully, pass a JSON schema, and still update the wrong customer. The value is present. The type is correct. The meaning is wrong because it came from the wrong source path. The failure that a schema cannot see Imagine an order webhook with two valid identifiers: { "webhook" : { "actor" : { "id" : "usr_101" } }, "order" : { "customer" : { "id" : "cus_9001" } }, "...
🌍 Dev.to

Constrained Decoding vs Post-hoc Validation: Production LLM Extraction Needs Both

Constrained decoding and post-hoc validation solve different problems. Constrained decoding is generation-time control: fewer malformed payloads less wrapper text better adherence to schema/tool shape Post-hoc validation is the trust boundary: coerce what is safe reject what is not attach confidence where routing matters A payload can be perfectly valid JSON and still be wrong for the workflow. That is why production extraction systems need both. This is also...
One line of math froze my AI agent forever. The timeout watched and did nothing.
🌍 Dev.to

One line of math froze my AI agent forever. The timeout watched and did nothing.

This is my second entry for the DEV x Sentry Bug Smash challenge. Entry #1 was a crash with a confusing error message . This one is the opposite and it is scarier: no crash, no message, no event. Just silence. The bug that sends you nothing smolagents runs LLM-generated Python in a sandboxed executor with a timeout. Issue #2473 claims that one line of model-generated math defeats it completely: from smolagents.local_python_executor import LocalPythonExecutor ...
🌍 Dev.to

You Deleted It. The Crawler Still Sees It.

I removed two listings from a directory site I run. Not for a small reason — they failed the one promise the directory exists to make. The kind of removal you do immediately and double-check. I did double-check. Flipped the flag in the database, loaded the site, watched the queries filter them out. Gone from the city pages, gone from the main directory. Removal confirmed. A day later, both were still on the site. Not on some cached page or stale CDN edge. In the current, freshly deploye...
HackerNews

Goodbye, and Thanks for All the Bikesheds

Article URL: https://queue.acm.org/detail.cfm?id=3818307 Comments URL: https://news.ycombinator.com/item?id=48960155 Points: 206 # Comments: 207
HackerNews

The Kimi K3 Moment

Article URL: https://stephen.bochinski.dev/blog/2026/07/18/the-kimi-k3-moment/ Comments URL: https://news.ycombinator.com/item?id=48960218 Points: 350 # Comments: 381
HackerNews

Deepsec

Article URL: https://github.com/vercel-labs/deepsec Comments URL: https://news.ycombinator.com/item?id=48964215 Points: 22 # Comments: 0
HackerNews

Scrying the AMD GFX1250 LLVM Tea Leaves

Article URL: https://chipsandcheese.com/p/scrying-the-amd-gfx1250-llvm-tea Comments URL: https://news.ycombinator.com/item?id=48965161 Points: 16 # Comments: 0
🌍 TechCrunch

Patreon stops asking AI bots not to scrape — and starts blocking them

Patreon is strengthening its defenses against AI scraping by working with Cloudflare to block bots that train AI models on creators’ content without permission. The move marks a shift away from relying on websites using robots.txt alone to actively block unauthorized AI training.