标签：#AR

Manually converting invoice data from PDF or Excel files into CSV format is one of the most time-consuming tasks in accounting and data management workflows. It often involves repetitive copy-pasting, formatting adjustments, and a high risk of human error. In many real-world scenarios, invoices arrive in different formats such as PDF, XLS, XLSX, or even HTML. Handling them individually can slow down reporting pipelines and create inconsistencies in structured data storage. The Problem with Manual Conversion Traditional invoice processing usually involves: Extracting line items manually from PDFs Reformatting Excel sheets for database compatibility Fixing inconsistencies in columns and values Rechecking for missing or misaligned data As invoice volume increases, these tasks quickly become inefficient and error-prone. Automated Approach to Invoice Conversion A more efficient approach is using tools that automatically parse invoice documents and convert them into structured CSV format. These tools typically: Read multiple file formats (PDF, XLS, XLSX, HTML) Detect table structures and line items Normalize data into rows and columns Export clean CSV files ready for spreadsheets or databases For example, uploading a multi-page invoice PDF can result in fully structured rows representing each item, without manual formatting adjustments. Why CSV Output Matters CSV remains one of the most widely used formats for: Accounting software imports Database ingestion Data analysis workflows Spreadsheet processing Having clean CSV output ensures compatibility across systems and reduces preprocessing work. Practical Impact Automating invoice-to-CSV conversion helps reduce: Repetitive manual data entry Formatting inconsistencies Processing time for bulk invoices It also improves accuracy when handling large datasets. Closing Note As data-driven workflows become more common in finance and operations, automating repetitive tasks like invoice conversion can significantly improve efficien

Most agents drag their entire past into every turn. A better default: keep a thin index of what was said hot, and fetch only the few turns you actually need — intact, on demand. Code: github.com/NirajPandey05/jit_context There is a quiet assumption baked into how most agents handle memory: that more context is safer than less. If the model might need something, put it in the window. The conversation grows, every prior turn rides along on every new request, and we trust the model to find the part that matters. That assumption breaks twice. It breaks on cost , because an agent loop re-sends its whole window on every step — a hundred stale turns aren't paid for once, they're paid for on turn 101, 102, and every step after. And it breaks on quality , because models don't read a long window evenly. Relevant facts buried in the middle get underweighted; irrelevant bulk competes for attention with the thing that actually answers the question. Past a point, a bigger context produces a worse answer, not just a costlier one. So the interesting question isn't "how do we fit more in?" It's "how do we keep the window small and dense without losing the one old turn that matters?" This post is the design we built around that question — for the specific case of long conversation history — plus the benchmark we used to keep ourselves honest. 01 · The mechanism: a hot index over a cold store The design borrows directly from how computers have always managed memory that doesn't fit: a small fast tier that's always present, a large slow tier that holds the bulk, and a rule for moving things between them. Virtual memory pages between RAM and disk. We page between the context window and an external store — for attention instead of address space. Concretely, there are two tiers. The cold store holds every turn at full fidelity, keyed by id — nothing is thrown away. The hot index holds one compact entry per turn: a short summary, a little metadata (entities, whether the turn recorded a dec

French serial entrepreneur and open-source legend Jean-Baptiste Kempf has been building Kyber, an infrastructure layer to control remote devices in real time.

AI isn't coming for your job. It already showed up, merged its first PR, and doesn't need a code review. The question developers keep dancing around — but rarely say out loud — is this: If GitHub Copilot, Cursor, and Claude can do what a junior dev does in a fraction of the time, what happens to junior devs? And more uncomfortably: what happens to mid-level devs in three years? The Uncomfortable Data Points This isn't speculation. It's already showing up in hiring data. Entry-level developer roles are contracting. Stanford's Digital Economy Lab (2025) found measurable decline in entry-level employment in AI-exposed roles — and software development is one of the most exposed. One senior dev + AI tools = the output of a small team. Brynjolfsson, Li & Raymond (NBER, 2023) showed generative AI productivity gains that compress what used to require multiple headcount into one. Goldman Sachs (2023) estimated significant white-collar labour market exposure — knowledge workers, not factory workers, are the primary target this time. This isn't the loom replacing weavers. It's the IDE replacing the person using the IDE. The Counter-Argument (And It's Not Weak) Here's where it gets interesting — because the doomsayer take isn't the whole story either. Every major technology wave destroyed jobs and created more than anyone predicted: The ATM didn't eliminate bank tellers — it lowered branch costs, banks opened more branches, teller roles increased for a decade The spreadsheet didn't kill accountants — it created an entire industry of financial analysts The internet didn't destroy publishing — it exploded the number of people who could publish The argument: AI raises developer productivity so dramatically that it expands the total addressable market for software. More products get built. More tools get created. More companies can afford to build what previously required a $500k engineering team. More demand for developers, not less. Where It Gets Complicated for Devs Specifically

If you've been following the AI market lately, you already know the headline numbers: DeepSeek V4 costs about 3% of what GPT-4o charges per token. GLM-4 runs benchmarks competitive with GPT-4 at roughly one-twentieth the price. Qwen delivers multilingual performance that rivals Claude for a rounding error in your cloud bill. The spreadsheets look incredible. The problem is actually using these models. Signing up for each provider means navigating Chinese-language dashboards, topping up separate wallets, managing six different API key formats, and dealing with SDKs that don't follow any consistent convention. Most developers give up after the second integration. That friction is why, despite the economics being objectively absurd in 2026, most teams still default to a single Western provider and eat the cost. AIWave exists to kill that friction. One API key. One endpoint. Fifty-plus models across eight Chinese labs, all speaking standard OpenAI-compatible format. Zero code changes to switch between DeepSeek, GLM, Qwen, MiniMax, and everything else. This post covers how the platform works under the hood, what the request lifecycle looks like, and how to integrate it in any language that can speak HTTP. The Fragmentation Problem, Quantified Before getting into the solution, here's what the Chinese LLM landscape actually looks like as of June 2026: Provider Flagship Model API Format Auth Method SDK Language DeepSeek V4-Pro Custom (DS format) Bearer token + signature Python, JS Zhipu GLM-4.5 OpenAI-compatible-ish JWT with expiry Python, Java Alibaba Qwen-3-Max DashScope (Alibaba) AK/SK + HMAC Python, Java, Go MiniMax MiniMax-Text-01 Custom REST API Key + Group ID Python Moonshot Kimi-K2 OpenAI-compatible API Key Python, JS Baidu ERNIE 4.5 Qianfan (Baidu) OAuth 2.0 Client Cred Python ByteDance Doubao-Pro Ark (Volcengine) IAM AK/SK + SigV4 Python, Go 01.AI Yi-Lightning OpenAI-compatible API Key Python Eight providers, seven different authentication schemes, four distinct A

Supervised vs. Unsupervised Machine Learning: How to Choose the Right Approach Supervised learning trains a model on data that's already labeled with the correct answer, so it learns to predict outcomes for new, unseen examples. Unsupervised learning works on unlabeled data and finds patterns or groupings on its own, without being told what the "right answer" looks like. Use supervised learning when you have historical examples of the outcome you want to predict; use unsupervised learning when you're trying to discover structure in data you don't yet understand. That's the short version. Here's what it actually means in practice, and how to know which one your project needs. What is supervised learning? In supervised learning, every training example comes with a label — the "correct answer" the model is trying to learn to predict. Feed a model thousands of emails, each tagged "spam" or "not spam," and it learns the patterns that separate the two. Once trained, it can label emails it's never seen before. The defining trait: you already know the outcome for your training data. You're not asking the model to discover something new — you're asking it to learn a pattern well enough to apply it to fresh cases. Common supervised tasks: Classification — sorting things into categories (spam vs. not spam, fraudulent vs. legitimate transaction) Regression — predicting a number (home price, next month's revenue) What is unsupervised learning? Unsupervised learning gets raw, unlabeled data and is asked to find structure in it — without anyone telling it what to look for. There's no "correct answer" to check against during training. The defining trait: you don't know the outcome in advance — you're trying to find it. A retailer might feed customer purchase histories into an unsupervised model not because they have a label called "customer segment" already assigned, but because they want the model to discover natural groupings on its own. Common unsupervised tasks: Clustering — gr

If you've been following the AI market lately, you already know the headline numbers: DeepSeek V4 costs about 3% of what GPT-4o charges per token. GLM-4 runs benchmarks competitive with GPT-4 at roughly one-twentieth the price. Qwen delivers multilingual performance that rivals Claude for a rounding error in your cloud bill. The spreadsheets look incredible. The problem is actually using these models. Signing up for each provider means navigating Chinese-language dashboards, topping up separate wallets, managing six different API key formats, and dealing with SDKs that don't follow any consistent convention. Most developers give up after the second integration. That friction is why, despite the economics being objectively absurd in 2026, most teams still default to a single Western provider and eat the cost. AIWave exists to kill that friction. One API key. One endpoint. Fifty-plus models across eight Chinese labs, all speaking standard OpenAI-compatible format. Zero code changes to switch between DeepSeek, GLM, Qwen, MiniMax, and everything else. This post covers how the platform works under the hood, what the request lifecycle looks like, and how to integrate it in any language that can speak HTTP. The Fragmentation Problem, Quantified Before getting into the solution, here's what the Chinese LLM landscape actually looks like as of June 2026: Provider Flagship Model API Format Auth Method SDK Language DeepSeek V4-Pro Custom (DS format) Bearer token + signature Python, JS Zhipu GLM-4.5 OpenAI-compatible-ish JWT with expiry Python, Java Alibaba Qwen-3-Max DashScope (Alibaba) AK/SK + HMAC Python, Java, Go MiniMax MiniMax-Text-01 Custom REST API Key + Group ID Python Moonshot Kimi-K2 OpenAI-compatible API Key Python, JS Baidu ERNIE 4.5 Qianfan (Baidu) OAuth 2.0 Client Cred Python ByteDance Doubao-Pro Ark (Volcengine) IAM AK/SK + SigV4 Python, Go 01.AI Yi-Lightning OpenAI-compatible API Key Python Eight providers, seven different authentication schemes, four distinct A

Anthropic’s Fable/Mythos shutdown is the first real model export-control shock The important AI story this week is not just that Anthropic launched bigger Claude models. It is that the US government then told Anthropic to switch two of them off for foreign nationals — and Anthropic says the practical answer was to disable them for customers while it works through compliance. That is a very different kind of platform risk than rate limits or pricing changes. If you are building on frontier models, model access can now move because of export-control decisions, safety claims, and geopolitical pressure. What happened Anthropic announced Claude Fable 5 and Claude Mythos 5 on June 9. Fable 5 was described as Anthropic’s most capable generally available model, with stronger performance across software engineering, knowledge work, vision, scientific research, and longer complex tasks. Mythos 5 was positioned above that: an upgrade to Claude Mythos Preview, with Anthropic calling out cyber-defence and life-sciences use cases. Three days later, Anthropic published a blunt update: the US government had issued an export-control directive requiring Anthropic to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States — including foreign-national Anthropic employees. Anthropic said the order arrived at 5:21pm ET on June 12, did not include detailed specifics, and that its understanding was that the government believed it had become aware of a jailbreaking method for Fable 5. Anthropic said access to other models was not affected, but the “net effect” was that it had to abruptly disable Fable 5 and Mythos 5 for customers to ensure compliance. Al Jazeera’s follow-up on June 19 frames the downstream effect clearly: allied countries and companies are now being forced to think harder about dependence on US frontier-model access. It also reports that Anthropic had granted roughly 200 institutions across 15 countries access to Claud

For the last 30 years, stopping the flow of cybersecurity-related software has proven to be ineffective. It's unclear why it would work now with Anthropic’s cybersecurity model Mythos.

Every developer who has ever muttered "there is a bug in this" is repeating a word with a surprisingly literal origin. On September 9, 1947, the operators of the Harvard Mark II, an early electromechanical computer, traced a malfunction to its source and found something they did not expect: a moth wedged inside Relay #70. They removed the insect, taped it into the operations logbook, and wrote a now-famous line beside it: "First actual case of bug being found." That page, moth and all, survives today in the collection of the Smithsonian's National Museum of American History. It is one of the best-loved stories in computing, and like most good stories it is a little more complicated than the popular version. Worth getting right, because the discipline it gave us is the same one behind every connected device we build. What actually happened in 1947 The Mark II was a room-sized machine built from relays, switches, and thousands of moving parts. When a moth flew into one of those relays, it physically interfered with the contacts and caused a fault. The technicians who found it had a sense of humor: calling it the "first actual case of bug being found" was a joke precisely because engineers had already been using "bug" for years to describe mysterious faults in machinery. Thomas Edison used the term in his notebooks back in the 1870s. So the 1947 moth did not invent the word "bug." What it did was give the term a perfect, photographable origin story, and it cemented the companion word that really matters: debugging. The act of removing that moth was, quite literally, de-bugging the computer. The Grace Hopper connection The story is almost always told with Grace Hopper at its center, and that deserves a small correction. Hopper, a pioneering computer scientist who later helped develop COBOL, was part of the Mark II team in 1947, but the evidence suggests she did not personally find the moth or write the logbook entry. What she did do was tell the story, brilliantly and o

What’s the most cliche possible gift you can give a relative? A digital photo frame, displaying a rotating slideshow of family photos. Now Aura has completely refreshed this product space with its gorgeous Aura Ink frame, which uses e-ink to create a display that doesn’t even look digital. Digital frames have always been so popular […]

Synthetic monitoring tools all promise the same thing — catch the broken checkout before your users do — and then bill you in seven different ways for it. The hard part of choosing one is not the feature checklist; it is predicting what you will actually pay when a single browser check running every 30 seconds from three regions turns into 259,200 runs a month. We compared seven synthetic monitoring tools on what separates them in practice: browser engine and fidelity, how you author checks (code, recorder, or AI), location coverage, alerting and on-call, failure forensics, and — the one that surprises teams — the pricing model. Every price below was verified against official pricing pages in June 2026. For the concepts behind these tools, start with what synthetic monitoring is . TL;DR comparison Tool Best for Browser engine Authoring Pricing model Browser price Checkly Code-first teams running Playwright suites Chromium (+ suite) Code (TypeScript) Per-run, 3 separate bills ~$4–6.50 / 1k Datadog Enterprises that want APM correlation Chrome/FF/Edge Recorder + code Per-run × freq × locations ~$12–18 / 1k Grafana Cloud / k6 OSS-leaning teams, best free tier Chromium (k6) Code (k6) + convert Per-execution ~$50 / 10k Better Stack Bundled monitoring + on-call Chromium Code + codegen paste Per-minute + per-seat ~$1 / 100 PW-min New Relic Broad type matrix + compliance Selenium (Chrome/FF) No-code step + code Per-check + seats + ingest ~$50 / 10k Sematext Predictable per-monitor pricing Chromium Code Per-monitor / month ~$7 / browser monitor Site24x7 No-code recorder + many locations Chrome/FF Recorder Pooled "advanced checks" ~$10 / 10k runs How we evaluated Real synthetic monitoring is more than a scheduled ping, so we scored each tool on six dimensions. Browser fidelity : does it run a modern engine (Playwright/Chromium) or older Selenium, and how faithfully does it reproduce a real user? Authoring mode : can you write checks as code, record them point-and-click, or gen

The short answer is: sometimes, but it's complicated.

Stop Fighting Scikit-Learn Pipelines: How Metadata Routing Fixes Sample Weights & Groups A couple of months ago, I stumbled upon this video by Vincent D. Warmerdam about metadata routing in scikit-learn. I'll be honest, I had no idea what "metadata routing" even meant, but Vincent's explanation completely changed how I think about building ML pipelines. The video showed me that one of the most frustrating problems in scikit-learn; passing sample weights and groups through complex pipelines finally had an elegant solution. It piqued my curiosity enough that I dove deep into the feature, tested it extensively, and honestly, I was surprised by how little coverage this gets in technical blogs and articles. So I figured, why not write about it myself and share what I learned? If you've ever struggled with imbalanced datasets, grouped cross-validation, or just wanted to pass custom information through your pipelines, this article is for you. Let's start from the very beginning. What is "Metadata" in Machine Learning? Let's start with a concrete example. You're building a credit card fraud detection model with this data: # Your training data X = transaction_features # Amount, merchant, time, location, etc. y = is_fraud # 0 = legitimate, 1 = fraud # But you also have additional information: sample_weights = [ 1.0 , 1.0 , 10.0 , 1.0 , ...] # Fraud transactions weighted 10x customer_ids = [ 101 , 102 , 101 , 103 , ...] # Which customer made each transaction Metadata is the "extra information" beyond your features (X) and labels (y): sample_weight : How important is each transaction? (Fraud = 10x more important) groups : Which customer does each transaction belong to? (For proper cross-validation) Custom metadata : Transaction timestamps, confidence scores, data quality flags, etc. Why Metadata Matters: The Credit Card Fraud Problem Imagine you're building a fraud detection system for a financial company. You have: Imbalanced data : 99% legitimate transactions, 1% fraudulent T

Whether you are building SaaS, running an agency, some shop, and struggling with your hard work - you have my highest respect. The ups and downs of working on your own is a killer man, it feels great on some days and just crazy on others. It's not for every one - if someone told me about this side of building startups 12 years ago (when I started my agency), I would have given it a bit more thought tbh. Keep building and shipping, and hopefully all your work will be rewarded in a way that feels rewarding to you.

Smart lighting company Philips Hue has launched its first wired wall modules. Installed behind existing wall switches, the new devices bring non-smart lights into the Hue ecosystem for the first time. Hue also announced new Play table and floor lamps that are more affordable versions of its Signe series, along with upgrades to its E14 […]

Fusion startups have raised $7.1 billion to date, with the majority of it going to a handful of companies.

A finance bot kept issuing refunds in a loop because nobody built a way to stop it. Clean code. Sound logic. No off switch. A small bug became a long night. Here is the opinion most teams do not want to hear. Building the agent is the easy 80 percent. That off switch is the 20 percent that decides whether you can ship it at all. We celebrate the wrong milestone. Picture the demo where the agent books the meeting, writes the email, updates the record. That part is genuinely fun to build and genuinely easy now. Harder is the boring question nobody claps for. What happens when it is wrong, fast, and confident. An AI agent is not a chatbot. It takes actions in the real world. It spends money, deletes rows, messages real people, moves files. Wrong answers in a chat are annoying. A wrong action at machine speed is an incident with your name on it. So before features, I build the stop. One real kill switch is not a single button. Think of it as a small set of bounds that live from the first version. A spend ceiling, so a retry loop cannot drain the account A blast radius limit, so one task can never touch more than it should A human gate on anything irreversible, so the agent proposes and a person commits A global stop that halts everything in one move, with no redeploy None of that is glamorous. All of it is what lets you sleep at night. Teams skip this for a reason that feels rational in the moment. Bounds feel like negative work. They never show up in the demo. Your agent runs fine without them right up until the one time it does not, and that one time is the only time anyone remembers. Here is the reframe that changed how I build. Treat the stop as the feature that makes an agent shippable. Bolt it on at the end and you have already shipped a liability that happens to pass the demo. Honest about the trade-off. Bounds slow you down. You will watch the agent pause for an approval it could technically have skipped, and it will feel like friction. That friction is the pric

Durante mucho tiempo asocié el desarrollo de software con programar funcionalidades: crear entidades, armar controladores, conectar una base de datos, validar formularios y hacer que una aplicación responda correctamente. Sin embargo, durante el Trabajo Final de la asignatura Desarrollo de Aplicaciones Web , entendí que programar es solo una parte del problema. El verdadero desafío aparece antes de escribir código: decidir qué arquitectura conviene, por qué conviene, cuánto cuesta, qué riesgos resuelve y qué complejidad agrega. El trabajo consistió en diseñar un sistema de gestión clínica que comenzaba como un MVP para una única clínica y evolucionaba progresivamente hacia una plataforma SaaS multi-tenant . Aunque fue un proyecto académico, el ejercicio nos obligó a pensar como si estuviéramos tomando decisiones técnicas en un contexto real: con restricciones de negocio, costos, equipo, seguridad, datos sensibles y crecimiento futuro. La principal enseñanza fue: la mejor arquitectura es la que responde mejor al momento del producto . El primer desafío: no sobrediseñar desde el inicio Cuando empezamos a pensar el sistema, la tentación era ir directamente a una arquitectura compleja: microservicios, eventos, colas, Kubernetes, múltiples bases de datos y despliegues independientes. Pero al analizar el escenario inicial, esa decisión no tenía sentido. El sistema comenzaba para una sola clínica, con un presupuesto reducido y con requisitos todavía en etapa de validación. En ese contexto, arrancar con microservicios hubiera agregado más problemas que beneficios: comunicación entre servicios, contratos, versionado, observabilidad distribuida, debugging más difícil y mayor costo de infraestructura. Por eso, una de las decisiones más importantes fue comenzar con una arquitectura en capas , desplegada como un único proceso. Esta elección permitió separar responsabilidades sin asumir desde el principio la complejidad de un sistema distribuido. La capa de presentación se encarg

Hey dev.to community! 👋 Over the last few months, juggling my roles as a Project Manager, Scrum Master, and lead for QA, Support, and Documentation has been a wild ride. The sheer speed of "vibe coding"—a paradigm shift where the primary role of the developer transitions from manual code construction to high-level intent orchestration—is incredible. Tools like Cursor, Replit Agent, and Google Antigravity allow us to scaffold entire microservices in minutes. However, while this transition offers unprecedented generative velocity, it introduces systemic risks concerning architectural integrity, long-term maintainability, and security. That’s why I’m sharing Cronos (Version 2.0) : a new, strategic methodology I’ve formalized for human-validated vibe coding and agentic software engineering. Real-World Testing & The Multi-Track Approach We have been rigorously testing this framework with our team over the last three months. The empirical results have been fantastic, tracking closely with the framework's theoretical efficiency models to deliver an almost 4x gain in productivity. To maintain production stability while achieving this speed, we adopted a multi-track approach. We continue to use standard Scrum for our maintenance track, which handles smaller tasks, support requests, and standard bug fixes. Meanwhile, Cronos is deployed exclusively for our parallel feature track, tackling larger Epics and new feature development. This ensures production stability does not stall innovation velocity. What is Cronos? The genesis of Cronos lies in the recognition that traditional Agile methodologies often fail to keep pace with the collapsed feedback loops of AI-driven development. In the current agentic era, the bottleneck has shifted from implementation to validation and strategic alignment. Cronos reconfigures the software development lifecycle (SDLC) around one-week "Cycles". Each cycle is a burst of high-intensity, AI-augmented creation coupled with a fixed duration of human