标签：#AI

Three weeks ago, one of the teams we work with had a checkout outage. The root cause a malformed...

Microsoft announced the general availability of Microsoft Discovery, its Azure-based platform for deploying autonomous AI agent teams in scientific R&D. The platform powered the development of Majorana 2, a topological quantum chip with 1,000x reliability improvement and 20-second qubit lifetimes. Microsoft now targets a scalable quantum computer by 2029, halving its original timeline. By Steef-Jan Wiggers

Today, June 8th, InfoQ celebrates 20 years. This is not a comprehensive history, but a deliberately selective look at the technologies and practices InfoQ identified early, where they sit on the adoption curve in 2026, and how that curve may evolve over the next five to ten years. By InfoQ

Microsoft announced Logic Apps Automation at Build 2026, a new SKU at auto.azure.com packaging workflows, AI agents, knowledge services, and model access into a managed SaaS experience. Agents integrate via agent-loop orchestration, Foundry agents, and managed sandbox. Knowledge as a Service provides a fully managed RAG pipeline. By Steef-Jan Wiggers

As an Engineering Manager in a Platform team, I manage 10 engineers. I'm hiring more. I run weekly 1:1s, facilitate technical decision meetings, screen candidates, moderate retrospectives, and still need to keep up with the delivery of a platform spanning dozens of AWS accounts. Besides the lack of time to focus on technical problems, the technical part is not even the real challenge. The less obvious problem becoming an Engineering Manager is: the skills you need as an engineering manager are fundamentally different from those that made you a great engineer , and there's no compiler or unit test to tell you when you're doing them wrong. The feedback loop is absent or very slow (and when you realise that, your team has already gone silent or become dependent on you because you are the main input and the main bottleneck). Skills That Don't Come From Code As a senior or staff engineer, you develop communication skills gradually. You present ideas, challenge others respectfully, summarise outcomes, and identify owners. You participate in technical deep dives and put candidates at ease while probing technical depth. These are valuable skills, and a good IC develops them over the years. But unless you start behaving like a brilliant jerk , they're secondary - your technical depth is still what defines you. But as an EM, the game changes. You're not "the smartest person in the room" anymore, and increasingly, you shouldn't be. You still have a broad context from all those alignment meetings and roadmap syncs, but you lose contact with the codebase week by week. If your organisation has principals or staff engineers, you're not even close technically anymore. Your job is to give direction, create space for others to solve problems, and facilitate decisions, not to be the one with the answer. This is hard. Especially when you used to be the one with the answer. The urge to jump in doesn't disappear just because your title changed. And interviewing? Facilitation? Giving feed

Most "Bitcoin in DeFi" stories quietly route through a custodian or a wrapped representation. You send BTC somewhere, someone (or some bridge multisig) holds it, and you get an IOU on another chain. That works until the thing holding your BTC is the thing that fails. For an autonomous agent that has to post collateral against an obligation it can't babysit, "trust the custodian" is exactly the assumption we're trying to delete. This post is about the alternative: a BTC collateral vault where native Bitcoin backs an obligation on another chain, the release is gated by a hashlock, and the worst case is a refund — not a loss. It's one of the primitives underneath Hashlock's settlement layer. I'll walk through the timelock ordering that makes it safe, the Bitcoin script that enforces it, and the failure modes you design around. Honest status up front: this is signet-validated, not BTC mainnet . The problem in one sentence An agent wants to commit BTC as collateral backing an action on Ethereum — settling a forward, anchoring one leg of a multi-leg trade, guaranteeing a payout — such that the BTC is released to the counterparty only if the corresponding obligation on Ethereum is fulfilled, and returns to its owner if it isn't. No third party should ever be able to hold, freeze, or abscond with the BTC in between. That's a cross-chain conditional. Bitcoin can't read Ethereum state, and Ethereum can't read Bitcoin's. The only thing both chains can independently verify is a hash preimage. So the entire construction hangs on one shared secret. The shared secret, and why timelock order is the whole game Both legs lock to the same hash H = SHA256(s) . Whoever knows the preimage s can claim. The instant s is revealed on one chain to claim a coin, it's public, and the other party copies it to claim the other coin. That's the atomic part: one preimage unlocks both legs or neither. The danger isn't the hash. It's time . If both legs had the same expiry, the party who knows the sec

A story about knowledge extraction, Kafka consumer rebalance, and what happens when a company...

The way developers write software has genuinely changed. Not incrementally — fundamentally. A year...

Intro AI Avatar is a completely free app that lets your VRoid (VRM) 3D avatar animate in...

I Tried PewDiePie's Open-Source AI Workspace. It's Actually Good. Yes, that PewDiePie. Felix Kjellberg (110M YouTube subscribers) spent late 2025 building a home AI lab — 8 modified RTX 4090s, 256GB of VRAM, running on Arch Linux. He called it "The Swarm." He crashed it running 64 models in parallel. The web frontend he built for it? He open-sourced it. Called it Odysseus . It hit 59,000 GitHub stars fast. I dug into the code expecting a glorified Ollama wrapper. It's not. What it actually is Odysseus isn't just another chat UI. It bundles things no other self-hosted tool does in one place: Chat — local or cloud models (Ollama, vLLM, llama.cpp, OpenAI, OpenRouter, GitHub Copilot) Agent mode — shell, files, web, MCP tools, per-tool toggles Cookbook — scans your GPU, recommends models that actually fit, downloads and serves them in one click Deep Research — multi-step web research that writes you a cited report Email — IMAP/SMTP with AI triage, auto-tagging, draft replies Calendar — CalDAV sync with Radicale, Nextcloud, Apple, Fastmail Memory — persistent, evolving across all your conversations No cloud account. No telemetry. MIT license. Everything lives in your data/ folder. The Cookbook is the standout feature Every other self-hosted UI assumes you already know what model to run. Odysseus doesn't. It scans your hardware, scores 270+ models against your actual VRAM, and gives you a one-click download-and-serve. It understands GGUF vs FP8 vs AWQ. It picks the right backend (vLLM, llama.cpp, Metal on Apple Silicon). Downloaded models persist in a volume — no re-downloading after container restarts. For someone who wants local AI but finds the ecosystem confusing, this is the most accessible on-ramp that currently exists. The code is better than the meme suggests The README has a little ASCII bear face. Don't let it fool you. The entry point app.py is 1,092 lines of real production thinking. A few things that stood out: The .env loader handles Windows BOM silently: loa

Quiet Defaults, DNSSEC Cracks, and Agents in the Data Plane I read the AWS Nitro V6 TCP timeout change twice before I believed it. Default went from 432,000 seconds to 350 seconds. Five days to six minutes. On the newest instance family. Quietly, in release notes most people won't read until something breaks. That sort of set the tone for May. No flagship launch to anchor the month around. What there was a lot of: defaults moving in places vendor press releases don't celebrate. Post-quantum crypto pushing into campus boot chains. Every cloud vendor shipping some flavor of agentic-networking pattern. The .de TLD briefly breaking because of DNSSEC. None of it announced loudly. All of it the kind of thing that breaks production at 2am if you weren't paying attention. What Moved This Month Three things, fast. Post-quantum crypto left the VPN tunnel. Cisco's full-stack PQC for campus and branch is the next chapter after April's PQ IPsec story — boot, firmware signing, supply chain attestation, and transport-layer crypto all moving together. If your campus has mixed-vintage gear (which is basically everyone), this is multi-year partial coverage with no clean switchover. Agentic networking became a real category. Cloudflare's Town Lake / Skipper writeup and Claude Managed Agents , Palo Alto's Portkey-based unified AI Gateway , and AWS's Bedrock AgentCore connectivity patterns all dropped this month. The right question stopped being "can my agent reach the model" and became "what IAM blast radius does this agent have if it gets prompt-injected." DNSSEC had a rough month. The .de TLD broke briefly, the DNSSEC root key was rolled, and Cloudflare also debugged a QUIC CUBIC death spiral that was hiding in plain sight. The Internet's core had a louder month than usual, and not in a good way. 1. Agentic AI Is Now Actually A Networking Problem An agent in production isn't a fancy chatbot. It's a thing that calls APIs, reads logs, accesses SaaS data, and sometimes writes back to sy

After running AI coding agents in production for a while, one thing became clear: the failures aren't in the code the model writes. They're at the seams — git, CI, auth, the network. The boundaries with the outside world. The model itself is genuinely capable. It writes functions, writes tests, refactors. What breaks is everything around the work: pushing the result, waiting on CI, merging the PR, refreshing a token, calling another service. And the failures are often the kind a human would avoid without thinking. Here are five incidents we hit and fixed in Codens' Purple (the orchestration core) over the last few weeks. All real, with production task IDs and dates. Every fix is merged. There's a shared design lesson at the end that ties them together. Incident 1: a half-resolved merge nearly flooded a PR with 12,000 lines This was the scary one. A Purple task on opsguide-back opened a PR. I looked inside: +12,162 lines / 149 files changed, with literal <<<<<<< markers in 2 of them . The commit graph: e567ce67 (merge commit, "chore: Fix HYBRID_SEARCH...") ├ parent[0] = 0b069e5d (develop tip, +1468 commits over main) └ parent[1] = 2940de35 (the actual feature commit) What happened: in the fix step, the AI decided to git merge develop to backport some test fixes. The merge conflicted. The AI resolved it partially and drove git commit through anyway with markers still in the tree. What got pushed: develop's entire divergence plus unresolved conflict markers. If anyone had clicked merge, main would have been polluted by 1468 commits of develop drift in one shot. A human wouldn't do this. They wouldn't merge develop into a main-targeted PR in the first place, and if it conflicted they wouldn't commit until it was fully resolved. But the AI, optimizing locally to get one test passing, does it without hesitation. Fix: stop it at push time, in two layers A single git pre-push hook. This is where the AI's git push actually goes, so this is where the guard belongs. #!/bin/bas

Your agent nailed every test case. You shipped it. Within 48 hours, users report hallucinated outputs, silently dropped tool calls, and responses that bear zero resemblance to what worked on your machine. You reload the same prompt locally. It works perfectly. Welcome to the most predictable failure mode in AI engineering: the dev-to-prod gap. This is Crucible C01. We dissect the five failure modes that kill agents in production and give you the tools to catch them before your users do. The Idea (60 Seconds) Developers test agents in idealized conditions: deterministic inputs, warm context windows, generous API latency budgets, and sequential tool calls. Production exposes the opposite environment: cold starts strip context, rate limits compress timing, and parallel calls introduce race conditions. The agent that performed flawlessly at temperature 0 on a 2k-token context window collapses at temperature 0.7 on an 8k-token window. The five failure modes are temperature drift and context window overflow first; silent API errors and prompt drift follow; race conditions complete the set. Each one has a detection pattern and a fix, and this article delivers both plus the CLI tool to automate the detection. Why This Matters AI agent failures differ from traditional software failures in one critical way: they are stochastic. A web API either returns 200 or 500. An AI agent returns something that looks plausible 90% of the time and is catastrophically wrong 10% of the time. That 10% is invisible in manual testing and devastating in production. The economics compound fast because every failed agent interaction wastes tokens, and wasted tokens cost money. At scale, a subtly broken agent burns budget faster than a working one because it retries, loops, and rephrases instead of succeeding. A single temperature drift bug can double your API spend. Reliability is the differentiator. The market is flooding with AI wrappers. The ones that survive will be the ones that work consiste

I haven't written articles in quite a while and I recently decided to come back to it. At work I use AI daily, so a big part of my coding tasks are delegated to agents. I try not to do the same when it comes to writing text but I don't have a clear reason for that (or maybe I do and I don't want to admit it). I am sure people can take advantage of generating text with the help of AI but at the moment I feel like prompting would not save me any time and writing the text myself would be faster. What about you? Are you using AI to write your articles? Image credit - jessica olivella, on pexels

The Letter VCs Are Quietly Deleting from ARR Startups are reporting revenue they haven't earned yet. VCs know it. Investors are cheering anyway. We've seen this movie. You're evaluating an AI startup. The pitch deck shows $100 million in ARR. The growth curve is parabolic. The deck says they signed $100 million. What it doesn't say is that $70 million of that is "committed ARR": contracts signed but not yet invoiced, customers who haven't deployed yet, pilots that count toward the number if they convert. Subtract the gap and you've got $30 million in actual recognized revenue hiding under a $100 million headline. This is the ARR inflation playbook, and it's running at full speed right now. The Trick Has a Name "CARR" stands for Contracted or Committed ARR. It's a legitimate concept. In industries where revenue accrues slowly after contract signing (healthcare AI deployments, energy optimization platforms, multi-year enterprise integrations), the gap between signature and recognition can legitimately take months or years to close. Reporting CARR alongside ARR, properly labeled, is defensible. That's not what's happening. What's happening is simpler: founders strip the "C" and just call it ARR. One VC told TechCrunch the gap between CARR and actual ARR can run as high as 70% [1]. In some confirmed cases the spread is 3-5x. Another investor said flatly: "For sure they are reporting CARR as ARR" [1]. The article indicates that the investor community is not only aware, but many are actively complicit. The logic follows its own warped rationality. When one startup in a category inflates, the others have to follow to stay competitive for talent and headlines. "When one startup does it in a category, it is hard not to do it yourself just to keep up," as one investor put it [1]. Spellbook CEO Scott Stevenson, one of the few willing to call this in public, described the practice as a "huge scam," adding that major VC funds are not just watching it happen but actively supporti

LLM-powered Learning, Handwritten Digit Recognition, and AI Career Guidance Today's Highlights This week's top stories showcase practical AI applications: an LLM-powered tool for domain learning, a cloud-enhanced handwritten digit recognition system, and an AI-driven career guide. These projects demonstrate how AI frameworks are being applied to real-world workflows, from knowledge acquisition to personalized advice. Show HN: Lathe – Use LLMs to learn a new domain, not skip past it (Hacker News) Source: https://github.com/devenjarvis/lathe This project, Lathe, presents a novel approach to leveraging Large Language Models (LLMs) not just for quick answers, but for deep, structured learning within a new domain. Unlike traditional LLM interactions that might encourage skipping detailed research, Lathe aims to facilitate a more profound understanding by guiding users through a systematic learning process. It likely employs advanced retrieval augmentation generation (RAG) techniques, potentially combined with iterative prompting strategies and graph-based knowledge representation, to help users build a comprehensive knowledge base on a chosen topic. The framework focuses on transforming raw information into actionable insights and structured learning paths. This makes LLMs a powerful study aid, enabling domain experts or newcomers to grasp complex subjects more efficiently by providing tools for semantic search, concept mapping, and progressive knowledge acquisition, moving beyond simple question-answering into true assisted learning workflows. Comment: This is precisely what's needed for complex enterprise knowledge management – turning LLMs into an active learning partner, not just a summarizer. I'd explore how it structures knowledge graphs or progressive learning paths. Handwritten Digit Recognition System with Cloud and AI Enhancements (Dev.to Top) Source: https://dev.to/yohannesah/handwritten-digit-recognition-system-with-cloud-and-ai-enhancements-i4e This project

I'm starting a new project. It's the most ambitious thing I've attempted from a phone. The goal: an AI agent that controls a smartphone. It opens apps, navigates screens, taps buttons, types text, and completes multi-step tasks. All offline. All local. No cloud. This is Day 1 of a public build log. No fluff. Just what I'm building, how it works, and what breaks along the way. What I'm Building An autonomous AI agent that runs entirely on an Android phone. You give it a command in plain English: · "Open WhatsApp and message Mom I'll call later." · "Search for Kotlin jobs on Wellfound." · "Open my notes and summarize what I wrote yesterday." The agent parses the command, plans the steps, and executes them—opening apps, finding the right buttons, typing text, hitting send. No cloud. No API keys. Just a phone that acts on your behalf. The Stack Component Tool AI Brain Gemma 4 E4B (local, via Ollama) Runtime Termux (Linux on Android) Phone Control ADB + UI Automator Orchestration Python Why This Matters Most AI agents live in the cloud. They need internet, APIs, and someone else's server. A local agent that runs on a phone means: · Privacy: your data never leaves your device. · Offline: works even without internet. · Accessible: built for the device billions of people already own. The Hard Parts I Already See · The agent needs to "see" the screen to know where to tap. Text detection is doable. Image-based buttons are harder. · Multi-step tasks need verification. If one tap misses, the whole chain fails. · Android permissions. ADB requires developer mode. A user-facing version would need a workaround. What's Next · Day 2: Create the repo. Set up the project structure. Push the first working script. · Day 3: Get screen text detection working with OCR. · Day 4: Test a full 3-step task. This is Day 1. The repo goes live tomorrow. Follow along if you want to see something rare get built from scratch.

Launch week for a side project means the same chore every time: export the logo, open Figma, create artboards for LinkedIn (400×400 and 4200×700 cover), Google Business Profile (1200×1200 JPEG with a minimum file size or Google rejects it), YouTube (2560×1440 with a safe zone nobody remembers), Open Graph (1200×630), Instagram (circle crop), Discord, TikTok… Design tools are great at making brand assets. They are slow at batch-exporting exact platform specs — especially when encoding rules differ (PNG vs JPEG, max KB, min KB for GBP, center-right vs YouTube-safe positioning). So we built a small free tool: one SVG or PNG in → 21 platform files out. What it actually does Upload a logo. Pick categories (or take everything): 8 profile icons — mark-only, padded for circular crops (LinkedIn, X, GBP, Facebook, Instagram, YouTube, TikTok, Discord) 2 profile lockups — logo + wordmark for LinkedIn company and GBP 7 banners/covers — including LinkedIn 4200×700, YouTube safe-area channel art, Facebook cover under 100 KB target, Open Graph 1200×630 4 post sizes — Instagram square/portrait/story, Pinterest pin Outputs are PNG or JPEG per platform rules. White background. ZIP download. Why not Canva? Canva is excellent for templates and marketing creatives. For “I already have a logo, give me every official size,” you duplicate frames and export manually — often behind a paid tier for brand kits and bulk workflows. This tool does one job, free, no account: 👉 https://varnox.io/tools Longer write-up with platform gotchas (GBP min JPEG size, YouTube safe area, etc.): 👉 https://varnox.io/blog/free-social-media-logo-size-generator Privacy (because it’s your logo) SVG uploads sanitized before render Generated files expire after 30 minutes — nothing permanent stored No signup We use the same catalog internally when onboarding clients onto GBP + LinkedIn + OG tags. Shipping it publicly was easier than answering “what size is our Facebook cover again?” for the fifth time. Stack note (for

We're likely to see more price increases as the big AI companies plan to go public.

Head's true genius—and that of his character, Giles—lay in quietly filling in the gaps in every scene