标签：#disc

"Auto-provisioning" is one of those features that sounds boring on a spec sheet and turns out to be doing a surprising amount of work once you look underneath it. The pitch is simple: drop in a list of users and the system spins up hundreds of configured softphones across devices in minutes, no manual setup per user. The pitch is easy. The plumbing behind it is where the interesting problems live. I've been picking this apart lately and wanted to write down what actually has to happen for "provision 500 users in minutes" to be true. What manual provisioning looks like (so you can see what's being automated away) Setting up one softphone by hand is a checklist. You create the user on the platform, assign an extension and credentials, point the client at the right SIP server, configure the transport (TLS, ports), set codecs, wire up voicemail, maybe push notification tokens, contacts, feature flags. Then you do it on their device. For one user, fine, twenty minutes. For five hundred, that's a full-time job for a week, and every manual step is a chance to fat-finger a config and generate a support ticket later. Auto-provisioning exists to make that whole checklist happen without a human running it each time. The core problem: getting config to a device that isn't configured yet The central puzzle of provisioning is a chicken-and-egg one. You want to hand a device its configuration, but the device doesn't yet know who it is or where to get that config. So the whole flow is about bootstrapping identity and config onto a blank client safely. A few common approaches: Provisioning URL + credentials: The user (or an admin) enters a provisioning username/password, or the client is pointed at a provisioning server URL. The client authenticates, the server looks up who this is, and returns the full config bundle. Simple, works, but needs the initial credential to get entered somehow. QR / activation code: Instead of typing SIP settings, the user scans a code or enters a short a

Welcome to this week's Top 7, where the DEV editorial team handpicks their favorite posts from the...

AI coding agents leave you with 5-to-20-minute gaps all day. Five real ways I fill them — and the one "quick" habit that makes you the bottleneck.

5:00 PM. Laptop closed. Slack closed. Workday over. Except my brain didn't get the memo. I'm making...

I wanted to disagree with 'AI made me a worse reviewer' from Michael Amachree (@dev_michael) ....

Your Agent Doesn't Have a Reasoning Problem, It Has a Memory Problem Anannya Roy Chowdhury Anannya Roy Chowdhury Anannya Roy Chowdhury Follow Aug 24 Your Agent Doesn't Have a Reasoning Problem, It Has a Memory Problem # ai # agents # architecture # programming 11 reactions 1 comment 9 min read

Meme Monday! Today's cover image comes from the last thread . DEV is an inclusive space! Humor in poor taste will be downvoted by mods.

🚀 From FlipaClip to SitePoint: The Full Story of Kehinde Owolabi How a Nigerian teenager built a professional game engine with borrowed laptops, offline W3Schools, and pure determination. 🎮 Play the Game Try Limn Engine Live — Space Shooter Demo See what 4 years of determination built. This space shooter runs at 60 FPS on a Tecno Pop 4 with 1GB RAM. 📖 Introduction Every developer has an origin story. Some start with a fancy computer and a computer science degree. Others start with a flipbook app and a sister who trusted them with her phone. My name is Kehinde Owolabi . I'm 18 years old (born December 4, 2007), and I live in Lagos, Nigeria. I'm currently in PC103 at BYU Pathway, and I'm a member of The Church of Jesus Christ of Latter-day Saints. I built a 94/100 professional game engine called Limn Engine. It runs at 60 FPS on a Toshiba with 4GB RAM. It was published on SitePoint and ranked #3 among 2D JavaScript game engines. Nobody knew it was developed on a Chromebook, a borrowed Thinkpad (behind my sister's back), and a Toshiba that "hung like hell." That was the secret I kept for months. But that's only one part of this story. This is the full story of how I went from a button phone to a 94/100 game engine, from FlipaClip to SitePoint, from a boy who failed physics to a developer who built something that runs on a Tecno Pop 4. The one-line summary: "I'm Kehinde Owolabi, an 18-year-old developer from Lagos, Nigeria who went from FlipaClip to building a 94/100 game engine on borrowed laptops — and got published on SitePoint." 🎮🚀 🎨 The Beginning: FlipaClip and the Spark of Creativity Before I was a developer, I was an animator. I used FlipaClip — a simple animation app on mobile — to create flipbook-style animations. I loved bringing characters to life, frame by frame. I would spend hours drawing, tweaking, and watching my creations move. That creative spark stayed with me. I wanted to create interactive experiences. I wanted to build games. But I didn't know how.

We often focus on what someone says, but in the realm of clinical psychology, how they say it is often more revealing. Subtle changes in speech—a slight tremor (jitter), a slowing tempo, or a flattened pitch—can be early indicators of depression or anxiety long before a user explicitly voices their distress. In this tutorial, we are building Psycho-Acoustic , a high-performance monitoring tool that leverages the HuBERT model , HuggingFace Transformers , and Librosa to quantify emotional states from non-verbal acoustic features. Whether you're interested in speech sentiment analysis , mental health AI , or advanced audio processing , this guide covers the end-to-face-mic implementation. The Architecture of Sound 🏗️ To accurately detect mental health indicators, we can't just look at text. We need a multimodal approach that combines raw signal processing with deep learning representations. graph TD A[Raw Audio Input .wav] --> B[Librosa Preprocessing] B --> C{Feature Extraction} C --> D[Traditional Features: Jitter, Shimmer, Pitch] C --> E[Deep Learning: HuBERT Embeddings] D --> F[Feature Fusion Layer] E --> F F --> G[Classification Head: Anxiety/Depression/Neutral] G --> H[Quantified Mental Health Score] H --> I[Deployment via ONNX Runtime] Prerequisites To follow this advanced guide, you’ll need: Python 3.9+ Tech Stack : transformers , librosa , torch , onnxruntime A basic understanding of digital signal processing (DSP). Step 1: Extracting Non-Verbal Acoustic Features 🌊 Before hitting the neural network, we need to extract "Psycho-Acoustic" features. Depression is often characterized by "speech prosody" changes—specifically reduced pitch range and slower speaking rates. import librosa import numpy as np def extract_prosodic_features ( audio_path ): y , sr = librosa . load ( audio_path , sr = 16000 ) # 1. Fundamental Frequency (F0) - Pitch f0 , voiced_flag , voiced_probs = librosa . pyin ( y , fmin = librosa . note_to_hz ( ' C2 ' ), fmax = librosa . note_to_hz ( ' C7

security, #api, #cybersecurity, #discuss On July 30, 2026, my batch job finished 300,000 domain-to-company lookups. 39,847 of them (13.3%) resolved to defense contractors, military-adjacent parent companies, or headquarters within a few miles of named bases. I wasn't hunting for that. I was just trying to clean a CRM. The same day, lina published a post about hijacking e164.arpa zones and accidentally logging hundreds of thousands of phone calls to military bases. Different protocol, same smell: an infrastructure lookup that was supposed to be boring turned into a classified-adjacent data spill. That parallel is what made me sit down and write this. Here is the exact call I used, with the live response for github.com so you can see the shape of the data before I explain what went wrong. import requests , json , time # Full source notes: https://github.com/On13uka/company-info-api RAPIDAPI_KEY = " YOUR_RAPIDAPI_KEY " BASE = " https://company-info1.p.rapidapi.com " def lookup ( domain ): r = requests . get ( f " { BASE } /lookup?domain= { domain } " , headers = { " X-RapidAPI-Key " : RAPIDAPI_KEY , " X-RapidAPI-Host " : " company-info1.p.rapidapi.com " }, timeout = 20 ) return r . json () print ( json . dumps ( lookup ( " github.com " ), indent = 2 )) The response I got back looked like this. It is a cached sample from a real call — the endpoint was asleep when I drafted this, but the fields are exactly what the pipeline consumed. { "domain" : "github.com" , "company_name" : "GitHub Inc" , "wikipedia" : "GitHub is a developer platform..." , "ceo" : "Thomas Dohmke" , "founded" : "2008" , "headquarters" : "San Francisco, California" , "employees" : "3000+" , "parent_company" : "Microsoft" , "twitter" : "@github" , "github_org" : { "repos" : 200 , "stars" : 50000 , "followers" : 12000 }, "health_score" : 78 } The Finding I started the job because a sales team had 300,000 stale domain records and wanted company names, headcounts, and a rough health score for each. The pla

The dictionary defines nuance as a subtle distinction or variation. In practice, it's the difference...

TL;DR Welcome back to Dev Opportunity Radar. This is a weekly series where I share opportunities,...

On-Device Piano Autocomplete: A 125M Model That Actually Works Meta Description: Discover how a 125M parameter model autocompletes piano music entirely on-device. We break down the tech, real-world performance, and what it means for musicians. TL;DR A developer shared on Hacker News that they trained a 125M parameter language model to autocomplete piano melodies — entirely on-device, with no internet connection required. The project demonstrates that compact, efficient AI models can deliver genuinely useful musical assistance without cloud dependency. This article breaks down how it works, who it's for, and whether it's worth your attention as a musician, developer, or AI enthusiast. Key Takeaways A 125M parameter model can run real-time piano autocomplete on consumer hardware without a cloud backend The model was trained on MIDI data and treats music generation similarly to how LLMs treat text prediction On-device inference means zero latency from network calls, full privacy, and offline usability This project sits at an exciting intersection of music technology, edge AI, and creative tools The approach has real limitations — it's not replacing a composer, but it's a genuinely useful creative assistant Similar techniques are being adopted in DAW plugins and music education software as of mid-2026 What Is "Show HN: I Trained a 125M Model to Autocomplete Piano On-Device"? If you spend time on Hacker News, you've probably seen "Show HN" posts — a dedicated space where builders share projects they've made. This particular submission caught significant attention: a developer trained a 125-million-parameter neural network to autocomplete piano music, and crucially, it runs entirely on your local device . No API calls. No subscription. No sending your musical ideas to a remote server. Just a model sitting on your machine, listening to what you play, and suggesting what might come next. The concept sounds deceptively simple, but the execution involves some genuinely clever

👋👋👋👋 Looking back on your week -- what was something you're proud of? All wins count -- big or small...

It’s been almost 2 months on Dev.to, and I’ve learned a lot from this community. Not just knowledge,...

I run self-hosted Chatwoot as the WhatsApp inbox for a dozen or so small Israeli businesses. Two servers, a few thousand conversations a week, a drip-sequence engine bolted on the side. Chatwoot is good software. The self-hosting docs will get you to a running container. What they will not tell you is which failures actually happen at month six, when you have real customers and real volume. These five all bit me in production, and none of them looked like what they were. 1. Your disk fills from somewhere Postgres never sees I got a disk alert at 86 percent and immediately went looking at the database. That was the wrong place. DB (postgres): 680 MB chatwoot_storage_data: 17 GB Attachments live in ActiveStorage, on a Docker volume, not in Postgres. Every image, voice note, and PDF a customer sends is a file on disk, and none of it shows up when you check database size. If your monitoring watches the DB, it will report everything is fine right up until the container cannot write. The growth curve is a function of how many accounts you host, not how busy any one of them is. Mine sat at roughly 0.05 GB a month until I onboarded seven new businesses over two months, and then it hit 16 GB a month. Check the right volume: docker system df -v | grep chatwoot_storage_data 2. Forty-four percent of my outbound storage was duplicate files This is the part that surprised me. When I actually measured what was on that volume, almost half the outbound media was byte-identical copies of the same file. One 14.5 MB video was stored 48 separate times. One image was stored 325 times. Chatwoot creates a new blob and a new file on disk on every send, even when the bytes are identical. That is correct behavior for a chat app where every message owns its attachment. It becomes expensive the moment you have anything that fans one file out to many conversations. In my case it was not campaigns at all, it was the drip engine sending the same media to 48 separate conversations as ordinary outbo

What if deciding what to eat felt as exciting as winning a championship? It's 2 AM. You're hungry. You open your favorite food delivery app, convinced you'll order something in two minutes. Thirty minutes later, you're still scrolling. Pizza? Burger? Pasta? Fries? Momos? Ice cream? Suddenly, every option looks equally good, and now you're questioning your entire existence just because you wanted dinner. I have this problem almost every time I order food. So when I saw the DEV Challenge, I wanted to build something fun around this tiny but painfully relatable problem. Unfortunately, I couldn't finish it before the deadline, but I still wanted to share the idea because it's one of those projects that made me smile while building it. Meet Foodwars . Instead of endlessly scrolling through hundreds of dishes, why not let your favorite comfort foods battle each other until only one champion remains? What I Built We've all watched cooking shows like MasterChef and somehow turned into professional judges sitting comfortably on our sofas. "That steak is overcooked." "The sauce needed more balance." "I would've plated it differently." As if Gordon Ramsay personally asked for our opinion. Foodwars lets us finally put those imaginary judging skills to good use. Instead of comparing hundreds of dishes at once, the platform randomly pairs comfort foods against each other in head-to-head battles. You become the judge. Pick the winner, move on to the next matchup, and continue until one food survives the tournament. No endless scrolling. No decision fatigue. Just a series of fun, quick decisions that eventually crown your Ultimate Comfort Food . And once the champion is decided... Go order it. Or cook it. Either way, dinner has finally been decided. Demo comfort-foodwars.vercel.app Features of Foodwars Foodwars isn't just a random food picker. Every round is designed to make choosing food feel like a game instead of a chore. 1. Interactive Tournament Brackets Instead of presenting

🚀 Technical Briefing: This tutorial is part of our deep-dive series on Agentic Workflows at Gate of AI . For the full technical breakdown, interactive code sandbox, and the native Arabic translation, visit the original article here . <p>Tutorial</p> <h1>Design a Safer AI Incident Copilot for GCC Operations</h1> <p>An AI incident copilot can help an operations team turn approved engineering facts into a clearer draft for stakeholders. It should not be treated as an autonomous incident commander, a source of truth, or an automatic publishing system. This tutorial explains how to define a safe operating model before choosing a framework, model provider, deployment platform, or integration.</p> <h2>Why incident copilots need a security-first design</h2> <p>During an incident, teams work under pressure. They need to communicate what is happening, who may be affected, what mitigation is under way, and when the next update will arrive. These messages must be accurate, calm, and consistent. An AI assistant may help prepare a first draft, but it can also amplify mistakes if it is allowed to infer missing facts, read untrusted material, or publish messages without review.</p> <p>The available security research on Copilot-style systems is a direct reason to design cautiously. Researchers have demonstrated ways AI systems can be manipulated to provide false references to files, extract some private data, and bypass security protections. The same research describes proof-of-concept abuse that can turn an AI assistant into an automated spear-phishing mechanism after an attacker gains the necessary access. These are not minor quality issues. They show that an AI feature connected to organizational information can become a security boundary.</p> <p>For an incident copilot, the safest initial scope is deliberately narrow: accept a small set of verified facts supplied by an authorized incident lead, create a draft in a fixed communication format, and require a human to review and pub

Opinion: The Diff Is a Claim, the Probe Is the Proof A generated patch is a claim about how a system should behave, and a diff cannot verify that claim on its own. The only honest reviewer is the runtime itself, which means every AI-proposed change deserves a behavioral probe before a human spends attention on it. Free model access changes the economics of that review, because the verification loop no longer costs a developer's full attention or a paid compute budget. The practical implication is that a disposable server, such as the free server option in MonkeyCode, becomes the arbiter of whether a patch is even worth reading. Disclosure: This article was prepared as part of MonkeyCode's product outreach. Review sessions routinely burn forty minutes on a diff that a five-second HTTP probe would have rejected instantly, and that waste is now entirely avoidable. Why Line-by-Line Review Fails on AI Patches A human reviewer reads a diff as prose, searching for the author's intent, but an AI-generated patch has no reliable intent to recover. The model that wrote the change cannot explain why a specific flag was flipped, and the diff itself only records the surface edit. This is a fundamental mismatch between the review tool and the review question. The review question is not "what changed" but "does the system still behave correctly after this change." Runtime shape diffing answers the first question well, and I have argued before that shape is a useful gate, but shape alone misses semantic regressions. A service can keep the same endpoints, the same config keys, and the same file layout while silently returning wrong data. Behavioral probes close that gap because they test the contract between the service and its callers. A probe sends real requests, checks real responses, and records real state transitions, which is exactly the evidence a reviewer needs. This is why I take the position that the probe, not the diff, should be the primary review artifact. Treat Every Pa

You budgeted six months. It took fourteen. You wanted one thing; you got three things that almost do it. And somewhere between the first sprint and the final invoice, you stopped understanding what you were even paying for. If that sounds familiar, this is the breakdown no one gave you before you started. What custom software development actually means Custom software development is building software from the ground up for your specific business, not configuring Salesforce, not installing a plugin. You're solving a problem your operations have, the way your operations actually work. What trips people up: "custom" doesn't mean "built entirely from scratch." Good dev teams use frameworks, libraries, and third-party services. What's custom is the logic of how your data flows, how business rules are enforced, how users interact. Scope range is huge: Custom dev covers everything from a lightweight internal dashboard to a full-scale multi-tenant SaaS platform. This is why cost estimates vary so wildly. 3 things nobody tells you before you sign 1. Scope creep is almost always the client's fault "Users should be able to manage their accounts" sounds simple. It actually contains dozens of decisions: can they change their email? What verification is required? Can they delete their account? Each one is a feature. Each feature has a cost. The fix: Run a discovery phase (2–4 weeks) before writing a single line of production code. It costs money upfront. It saves far more mid-project. 2. The cheapest bid rarely wins long-term A $40k quote and a $180k quote for the same project both happen. The $40k team isn't lying; they're optimistic, underbidding to win work, or scoping something different. What actually happens: you hit $40k, and you're 40% done. Higher bids from experienced teams often include architecture planning, documentation, testing infrastructure, and post-launch support things the cheap bid omitted. These aren't extras. They're what make the software maintainable in t