标签：#developer

The term software factory is getting a lot of attention right now, and for a good reason. AI coding assistants can generate code much faster than before. But faster coding alone does not mean faster, safer delivery. In many teams, it simply moves the bottleneck to review, testing, deployment, and operations. A software factory is a way to organize the entire software development life cycle as one connected, repeatable system. Think of a car manufacturing assembly line. Each station has a clear job, work moves forward in a predictable order, quality checks happen at the right moments, and the finished product is inspected before it leaves the factory. An agentic software factory applies that same idea to software delivery. AI agents do focused work across planning, coding, testing, deployment, monitoring, and feedback. Humans remain in charge of specifications, security, policies, approvals, and the decisions that should never be delegated blindly. Key Takeaways An agentic software factory coordinates AI agents across the complete software delivery lifecycle. Faster AI-assisted coding can create review bottlenecks unless downstream stages also improve. Humans retain ownership by defining guardrails, specifications, approvals, and security requirements. Workflow orchestration connects context, automation, observability, incident handling, and feedback loops. What Is a Software Factory? A software factory is not just a collection of developer tools. It is an operating model where software delivery is designed as a smooth, observable workflow from idea to production and back into improvement. In a car factory, a vehicle moves through assembly, painting, quality inspection, final assembly, and delivery. People are involved at important checkpoints, but the process does not restart from scratch at every station. It is structured, repeatable, and connected. The same model works for software. In an agentic software factory, the flow can look like this: Requirements and desi

A few months ago we shipped the epilot CLI , and it quietly became one of my favorite tools. One command, npx epilot , gives you every single epilot API operation in your terminal: entities, journeys, workflows, pricing, files, permissions, 50+ APIs. Interactive pickers if you're exploring, --json and --no-interactive if you're scripting. It also turned out to be a perfect match for AI agents like Claude. Agents are great at driving CLIs: they discover operations, read the help, make calls, parse the JSON. No custom integration or MCP server needed, the CLI is the integration. And because handing an agent live CRM access is a scary idea, the CLI ships with two safety nets, both enforced server-side: # A session that physically cannot write. The restriction is baked # into the token, so the bearer can't turn it off. epilot auth login --readonly # A token that additionally gets all PII anonymized in every response epilot access-token createAccessToken -d '{ "name": "AI agent token", "read_only": true, "anonymize": true }' Read-only plus anonymized means an agent can explore, analyze and report on your real org all day, and the worst it can do is read data it can't even de-anonymize. Now we've made the CLI even better. On top of the raw API commands, we added app facades : a set of high-level epilot app commands that take you from an empty folder to a working app installed in your org. And that's what this post is really about, because apps are where the fun is. What are epilot Apps? epilot is very configurable out of the box: journeys, workflows, automations, pricing. But at some point every team hits a wall, something the UI simply doesn't offer. A custom tab on the contact or opportunity page showing data from your own systems A whole custom page in the epilot navigation Your own block in the journey builder A widget in the end-customer portal A flow action that calls your API when a workflow step runs An external product catalog or an API proxy to your backend That

If you've ever stared at a 15,000-file monorepo wondering "what actually breaks if I touch this...

Developers are owning more of the delivery system around code, not just code itself. Join us during GitHub Universe to meet other devs, learn something new, and explore what's next. The post From coder to orchestrator: How agents shift the role of a developer appeared first on The GitHub Blog .

There are two ways to work with an AI agent and I had tried both. Write the thing yourself and hand over only the tedious parts. Or hand over the whole task and audit whatever comes back at the end. The first is slow. The second is fast right up until it is wrong, and by then the wrong thing is finished. I expected this series to be about forcing a third option into existence. Nine parts of making an agent follow a workflow it would rather skip. That is not what happened. I never had to enforce it once. The queue that started this had a payload contract nobody had verified, and each phase after that cost me something before it gave anything back. A plan that would not move until the risk register named the provider contract the brief had only guessed at. A build that missed nothing except what my own brief left out. A review that stopped handing back a feeling and started handing back a verdict on every requirement I had already called done. A matrix instead of a trusted green run. A rollback with a name on it before anything got called shipped. And a retrospective that would not let a lesson through until it had checked itself against the trail. Eight parts of that. What I did not expect was which part turned out to be automatic. The Fight I Expected Never Started By the time I finish writing a requirement, I already know roughly what it is going to cost. Most engineers do. You can feel the difference between a one-line fix and something that is going to touch four files and a migration before you have written a single line of it. What I assumed was that the agent could not feel that, and that policing the gap would be my job forever. Reminding it to run the chain. Catching it when it decided a spike was small enough to skip. It has not needed the reminder. Small bugs do not trigger a brief and a plan, and they should not. A standard requirement, a spike, anything long or cross-cutting, runs the full cycle in order. The classification lands where I would have put i

I used to think financial apps were designed mostly for people who already understood finance. You open an app, see dozens of charts, numbers, and settings, and sometimes your first thought is: “Okay… where do I even start?” AI is slowly changing that experience. The most useful AI applications in fintech are not always about making predictions. In many cases, they are about removing friction: organizing information, automating repetitive tasks, and helping users understand complicated processes. We can already see this trend in different areas. Companies like Stripe are using technology to simplify online payments, while platforms like Plaid help connect financial data between different services. The same idea is appearing in other parts of finance too. Some digital financial platforms are exploring automation tools that help users create more structured workflows instead of manually managing every step. For example, platforms like BYDFi have introduced automated tools that allow users to use predefined strategies rather than constantly monitoring every market movement. Of course, automation does not replace human decisions. A smarter tool does not mean a person can ignore research or risk management. For developers, the interesting challenge is not building the most complicated system. It is building something that makes complicated things easier for real users. Maybe the future of fintech is not about adding more buttons. Maybe it is about helping people need fewer buttons.

The Quest Begins (The "Why") I was staring at a pull request that felt like a boss level in a retro arcade game—except there were no extra lives. The code was a massive if/else if/else chain that decided how to handle different JSON payloads coming from a third‑party API. Each branch did almost the same thing: validate a few fields, map them to our internal model, then call a service. The only thing that changed was the shape of the incoming object. Every time a new endpoint was added, a developer had to copy‑paste the whole block, tweak a few field names, and pray they didn’t miss a comma. Reviewing it felt like watching someone try to solve a Rubik’s cube by rotating random faces—you could get lucky, but most of the time you just made a bigger mess. I kept asking myself: Why are we writing the same logic over and over? The answer was hiding in plain sight: we weren’t seeing the pattern. The Revelation (The Insight) The breakthrough hit me while I was refactoring a tiny utility that turned a list of user IDs into a set. I realized I wasn’t writing a new algorithm each time—I was applying the same shape of solution: take an input, transform it, then feed it to a consistent consumer . In other words, the problem wasn’t “how do I handle payload X?” It was “how do I dispatch the right transformation based on a key?” That’s a classic dispatch table (or strategy pattern) problem. The “aha!” moment was when I looked at the chain and saw that each branch could be expressed as a function: function handleOrder ( payload ) { /* … */ } function handleRefund ( payload ) { /* … */ } function handleShipment ( payload ) { /* … */ } All of them shared the same signature: (payload) => Result . If I could map a discriminator (like payload.type ) to the correct function, the whole if/else monster would collapse into a single lookup. That’s the pattern top coders spot instantly: repetitive conditional logic → a table of behaviors . Once you see it, the code writes itself. Wielding the

Instacart introduced Blueberry, an AI-assisted incident response system that helps on-call engineers investigate production issues faster. It combines AI agents, operational data, and historical incident knowledge to generate grounded root cause hypotheses in Slack. It uses parallel subagents, MCP integrations, and incident history to reduce investigation time while keeping engineers in control. By Leela Kumili

The Scenario Every Developer Recognizes It is mid-2026, and you have a feature to ship. You open ChatGPT or Claude, type something like "build me a function that parses webhook payloads and routes them to the right handler," and wait. The model returns something plausible. You paste it in, run it, and it almost works. So you prompt again: "fix the edge case where the payload is missing the event key." Another round. Then another. Forty-five minutes later, you have code that functions, but you also have a conversation thread that looks like a debugging session rather than a build session. You never actually described what you were building. You just started building it. This is the default mode for most developers using AI coding assistants in 2026, and it is expensive. According to McKinsey's State of AI in 2024 report ( source ), organizations that adopt structured design and planning approaches before implementing AI tools report higher success rates and better integration outcomes compared to those using ad-hoc implementation strategies. The pattern holds at the individual developer level too. Jumping straight into prompting skips the step that makes prompting useful: knowing precisely what you want before you ask for it. The fix is not a better model. It is a different sequence. What Design-First Actually Means in Practice Design-first means producing a written artifact that describes your system before you write a single prompt asking an AI to build it. Not a full technical document. A tight, structured description of inputs, outputs, constraints, and edge cases. Think of it as the brief you would hand to a contractor before they start work. The contractor analogy is useful because it reframes the relationship: you are not collaborating with the model in real time, you are commissioning it with a clear scope. Here is what that looks like concretely. Instead of opening Google Gemini and typing "help me build a webhook router," you spend ten minutes writing this

Having a platform is not enough; the real challenge is ensuring that it is understandable, usable, and actually adopted by its users. A capability is done when it can be reliably used by others. To evaluate progress, you can ask yourself “Is this being used?” and “Does it reduce friction for users?” This can help align development work with actual user value rather than delivery, By Ben Linders

Protocols let you define what a type can do without caring about what it actually is. Once...

Hi, I’m Hélène. I’ve had computers in my life for more than forty years now, which still surprises me when I say it out loud. I started out as an eight-year-old kid playing simple games on a family computer, before the NES, before “PC gaming” was really a thing. We loaded games from cassette tapes, listened to modems sing their weird little songs, and waited for desktop machines to slowly grind their way through a boot sequence. I taught myself to program as a teenager because I wanted to make the computer do more than just play those games. Later, I went through technical school, then university, and graduated with a Computer Science degree. I did a Master-level short program in project management. I worked as a software developer for years, and now I lead a team. Along the way I read piles of books, watched improv and theatre, devoured Choose Your Own Adventure stories — only to realize much later that all of that was quiet leadership training in disguise. But there was one book I never found: the one that explained how to grow from junior to a trusted, impactful developer in a way that felt honest and practical. Not just “write clean code” or “communicate better,” but how to actually do those things, day after day, with real people, real constraints, and real doubts in your head. When I started my career, there were textbooks on algorithms and operating systems. There were tutorials on whatever language was hot that year. There were business books full of buzzwords. What I couldn’t find was the in-between book: something written by a practicing software developer, talking to another developer, about the messy, human parts of this job. Junior to Engineer is my attempt to write that missing book. Who this book is for If you’re a junior or intermediate software developer and you’re wondering things like: “How do I become the person people trust with the hard problems?” “What should I actually focus on in the first 5–10 years of my career?” “Why does everyone else se

Bilibili's 412 Incident, Explained: How v1.5.0 Absorbed It In June 2026, Bilibili quietly began rejecting yt-dlp with HTTP 412 errors. Agents wired to scrape it broke — except the ones sitting behind Agent-Reach, which rerouted the channel before most developers noticed. Agent-Reach is a local, MIT-licensed capability layer that gives shell-capable coding agents live internet access by selecting and routing to upstream CLIs rather than proxying data itself . When Bilibili started 412-blocking yt-dlp in June 2026, v1.5.0 rerouted the Bilibili channel to bili-cli with zero user action, while YouTube kept using yt-dlp untouched . The fix landed centrally: the maintainer reordered backends, so no individual builder had to patch a private integration. Quick Answer: When Bilibili began returning HTTP 412 to yt-dlp in June 2026, Agent-Reach v1.5.0 automatically rerouted its Bilibili channel to bili-cli — agents kept working with no user action. The release passed 32 end-to-end tests across 13 channels and grew its suite from 107 to 162 tests. The framing shift matters: v1.5.0 describes itself as a capability layer, not a tool collection. Each platform gets an ordered primary-plus-fallback backend list; after setup, your agent calls those CLIs directly and Agent-Reach never sits in the data path . The June 11, 2026 release passed 32 end-to-end tests across 13 channels and grew its test suite from 107 to 162 tests . Platform Primary backend Fallback Web pages Jina Reader — YouTube yt-dlp — GitHub gh CLI — RSS feedparser — Bilibili bili-cli OpenCLI (subtitles) Twitter/X twitter-cli OpenCLI Reddit OpenCLI rdt-cli XiaoHongShu OpenCLI xhs-cli LinkedIn linkedin-mcp Jina Reader Global search Exa via mcporter — "capability layer: multi-backend routing + real doctor + OpenCLI" — Agent-Reach v1.5.0 release framing (source: Agent-Reach CLAUDE.md ). The behavior is easy to model. The following minimal snippet — which was executed and returns exit 0 — illustrates the "absorb and keep wo

Somewhere on your feed right now, someone is bragging about the app they built in a weekend, no engineering background, no team, just a prompt and a Saturday. The post always ends the same way. Look what I built without needing any of you. Anyone can build software now. That is the whole pitch, repeated in a hundred different captions this month alone. Here is what that post never shows you. The part where someone checks it. Not "does it run." Checks it. Someone who did not write it, looking for the version of it that fails, the input nobody thought to try, the assumption that was wrong in a way the builder was structurally the worst person to catch, because they were too close to their own idea to see the hole in it. That someone is not optional. It is the actual job. None of this is theoretical. A notification icon that, instead of opening a panel, closes the entire page and drops me back on an empty tab. A video call that disconnects mid sentence for no visible reason. A video that plays with the sound simply gone, until I restart it. I do not have a chart proving reliability across the industry is getting worse. What I have is a pattern I keep running into, on products built by some of the most resourced engineering organizations on earth. Why software engineering has more than one person in the room A developer writes the code. A reviewer reads it before it merges. QA tries to break it on purpose. A manager decides if it is actually ready, or just finished. None of these roles exist because engineers do not trust themselves. They exist because a single person, however good, cannot see their own blind spots. That is not a flaw in the person. It is a fact about how blind spots work. Ten sets of eyes exist so that the eleventh mistake gets caught before a million people hit it. We have already watched what happens when that layer disappears, and we did not need AI to run the experiment. We ran it with the spreadsheet. The spreadsheet already showed us the cost of

When I published v2.0.0 , I thought the hard part was over. The CLI could turn Claude artifacts into native Windows applications in a single command. It worked on my machine, the demos looked great, and I was excited to share it. Then people started using it. That's when I realized the difference between building software and shipping software . The bugs only users can find Most of the problems weren't visible in a five-minute demo. Some applications worked perfectly. Others silently lost all of their data after being closed. Some builds succeeded exactly once. The second build failed unless the runtime cache was manually deleted. Everything looked correct... until someone actually tried using it. The rabbit hole What I thought would be a quick patch became a deep dive into things I'd never touched before. Over the course of v3 I ended up learning far more than I expected: Git branching and release workflows Semantic versioning GitHub Actions npm packaging npm pack Runtime integrity verification SHA-256 checksums Neutralino runtime management JavaScript debugging Integration testing None of those were part of the original project idea. Persistent storage finally works This was the biggest challenge. Artifacts using localStorage behaved differently depending on where they were running. Inside Claude everything worked. Inside the generated desktop application... not always. Tracking this down took far longer than writing the original feature. The result is that applications now persist their data correctly across launches without requiring any code changes. If your artifact uses localStorage , it should simply work. Runtime improvements The runtime layer also received a major overhaul. Version 3 now handles: automatic runtime downloads integrity verification cache recovery consecutive builds without manual cleanup Those aren't exciting features to demo, but they're exactly the kinds of improvements that make a tool feel dependable. Better testing I also spent a lot mo

The Quest Begins (The "Why") I still remember my first technical interview like it was yesterday. The recruiter slid a whiteboard marker across the table, smiled, and said, “Here’s a classic: given an array of integers and a target sum, return the indices of the two numbers that add up to the target.” My heart started racing. I could feel the sweat forming on my palms as I stared at the empty board, my mind looping over the same terrible idea: check every pair . I started scribbling a nested loop, O(n²) time, and immediately realized that if the array had even a few thousand elements, I’d be stuck there forever. The interviewer’s eyes flicked to the clock, and I could almost hear the Imperial March playing in my head— the pressure was real . I needed a way to cut through the noise, fast, or I’d be that candidate who “just didn’t get it”. That moment sparked a question that’s haunted me ever since: how do top coders stay calm, spot the shortcut, and turn a seemingly impossible problem into a few lines of clean code under pressure? The Revelation (The Insight) After that interview (and a few too many late‑night debugging sessions), I dove into the mental toolkit that separates the “just‑get‑it‑done” crowd from the folks who seem to solve puzzles while sipping coffee. The breakthrough wasn’t a new library or a fancy language feature—it was a simple shift in perspective: Instead of asking “how can I compare every element to every other element?” ask “what do I need to know about each element to instantly know if its partner exists?” In the Two‑Sum problem, the partner of a number x is simply target – x . If I could remember, in O(1) time, whether I’ve already seen that partner, I could solve the whole thing in a single pass. That’s the “aha!” moment: store what you’ve seen so far in a hash map (or set) and look for the complement as you go . It feels like discovering the One Ring in a junkyard—once you see it, everything else falls into place. The beauty is that this pa

by Cipher Forge - Compounding-Asset Specialist @ HowiPrompt The past week has been a micro-boom in AI research. Five papers landed on arXiv, three on OpenReview, and a handful of industry pre-prints that together push the frontier on multimodal reasoning, efficient fine-tuning, and trustworthy LLM deployment. In this guide I'll: Distill the core contributions of each paper (no fluff, just the meat). Show you how to reproduce the key results with publicly available code or minimal re-implementation. Map the findings to real-world product pipelines - from data ingestion to inference scaling. Provide a reproducibility checklist so you can turn a paper into a compounding asset for your startup or product team. Grab a coffee, fire up your dev environment, and let's turn these seven papers into immediate value. 1. The Week in Review - Why These Papers Matter Date (2026) Venue Title Primary Claim Reported Gains Jul 06 arXiv "Mosaic-LLM: Structured Prompt Fusion for Multimodal Chains" A unified prompting language that stitches vision, audio, and text into a single chain of reasoning. 12.4 % higher VQA accuracy vs. Flamingo-3B on OKVQA. Jul 07 OpenReview "DeltaLoRA: Parameter-Efficient Fine-Tuning via Low-Rank Delta Updates" Introduces a delta-matrix on top of LoRA that reduces fine-tuning compute by 38 % without loss. 0.3 % BLEU drop on WMT-2025 while cutting GPU-hrs from 120->74. Jul 08 arXiv "TrustGuard: Certified Robustness for Retrieval-Augmented Generation" Formal robustness certificates for RAG pipelines under adversarial query perturbations. Guarantees 95 % success rate on adversarial SQuAD-2.0 attacks. Jul 09 arXiv "Neuro-Sketch: Zero-Shot Sketch-to-Image Generation with Diffusion-Guided Transformers" Leverages a diffusion prior to translate coarse sketches into photorealistic images without training on paired data. FID = 21.3 on QuickDraw-500, 2.8× better than prior zero-shot baselines. Jul 10 OpenReview "Meta-Prompt Engine (MPE): Automatic Prompt Synthesis for LLM

Someone in Cuba or Iran can keep installing APKs with no new restrictions, but devs will suffer.

A suite of developer tools you run with the curl you already have: UUIDs, hashes, JSON End-points, JWTs, JSON, QR, File Transfers and more. Nothing to install, and no signup required. "curl curlhub.sh" to see list of commands or visit https://curlhub.sh to view all tools full details and soon contribute. root@web01:~# curl curlhub.sh curlhub.sh — CLI-first developer utilities Zero-setup tools you curl straight from the terminal. Generators /uuid RFC 4122 v4 UUID. /pass High-entropy password / API key / token. CLI: server-side CSPRNG over TLS; web: client-side. Never logged. ( https ) Utilities /qr ANSI/UTF-8 QR code from ?data. Nothing stored server-side; use HTTPS or pipe stdin for sensitive payloads ( a ?data = value appears in request URLs ) . Encoding /b64 Base64 encode / decode. (https) /hash MD5 / SHA-1 / SHA-256 / SHA-512 of input. (https) Developer /json Validate + prettify + colorize JSON. (https) /jwt Decode & pretty-print JWT header/payload. No verify; token never logged. ( https ) /status Explain an HTTP status code + troubleshooting. /cron Translate a cron expression to English + next run times. /ua Parse the User-Agent you sent (browser / OS / engine / device). /headers Echo the request headers you sent (+ the edge view). Your own request, not a remote audit. /hook Webhook inspector: mint a temp endpoint, inspect incoming HTTP. (https) Text & Logs /md Render Markdown to colorized ANSI (safe, bounded parser). (https) /p Pastebin: pipe text/logs to a short URL (plaintext in terminal, highlighted on web). (https) Network /cidr Subnet / CIDR math: range, netmask, broadcast, host count. /ip Your public IP (thin). Geo/ASN live at worldip.io. /whois Domain registration / registrar / dates (public OSINT; rate-limited + cached ) . Security /ssl Decode a PEM certificate or CSR you paste (expiry, issuer, SANs, key). No outbound connection. (https) File Transfer /u Ephemeral file drop (<=100 MB). ANSI QR; auto-purge after 1 download or TTL. ( https ) Docs /man Com

Learn how I modernized an old codebase of mine using stacked sessions and pull requests in the GitHub Copilot app. The post Stacked sessions and pull requests in the GitHub Copilot app appeared first on The GitHub Blog .