标签：#an

Docking stations expand what your laptop can do, and I’ve been testing the best of the best to see which you should buy.

From halting my doomscrolls to automating air-quality checks, Apple’s streamlined Shortcuts app is my favorite iOS 27 feature.

OVHcloud will raise prices from September, with 2026-edition gaming servers up 87 percent and other recent servers 40 to 59 percent. Founder Octave Klaba says memory cost six times more in June than a year earlier, as RAM suppliers shifted capacity toward high-bandwidth memory for AI. AWS, buying years ahead, has repriced one reserved GPU product. By Steef-Jan Wiggers

Every accessibility tool I have used reports violations like this: Images must have alternative text body > main > div:nth-child(2) > form > div.field > img That selector is correct. It is also useless. It describes the rendered DOM , and I do not write rendered DOM — I write templates. Somewhere in a few hundred .component.html files there is an <img> that produced it, and finding it is manual work: grep for img , get forty hits, open them one by one, compare surrounding markup until something matches. Multiply that by sixty violations and the scan stops being useful. Not because it is wrong, but because acting on it costs more than ignoring it. React solved this years ago If you write JSX, babel-plugin-transform-react-jsx-source puts a _debugSource on every element at build time — file, line, column. That is how React DevTools can jump you straight to source, and how error overlays point at the right line. Angular has no equivalent. The compiler knows the position of every element in every template: it has to, to report template errors. But nothing carries that knowledge into the DOM. So I built the bridge. parseTemplate hands you the positions @angular/compiler exports parseTemplate , the same entry point @angular-eslint uses. Give it a template string and you get an AST where every node carries a sourceSpan with byte offsets, lines and columns: import { parseTemplate } from ' @angular/compiler ' ; const parsed = parseTemplate ( source , filePath , { preserveWhitespaces : true }); // each element node has startSourceSpan.start.{offset,line,col} Two things to know immediately. The compiler counts lines and columns from zero , and every editor counts from one — so you add one, or every location you report is off by one in both axes and nobody trusts the tool again: line : span . start . line + 1 , // the compiler counts from zero, editors do not column : span . start . col + 1 , And preserveWhitespaces: true matters: without it the offsets you get back describe a t

How I built an FVG trading bot for OKX and made 99% of its signals useless on purpose If you trade crypto futures, you know the drill. You're staring at the chart at 3am because you're scared to blink and miss "the perfect entry". Or worse, you get in emotionally, chase a pump, and hand back all your profit in one bad night. I got liquidated once because my stop was at -5% and the liquidation price was at -2%. Price gapped straight through my stop. That's how this project started. I built FVG Killer , a bot that trades one setup only: the ICT Fair Value Gap, on OKX perpetuals. The repo is open-source: https://github.com/Xbs950812/okx_fvg_agent 1. What it trades FVG stands for Fair Value Gap, from the ICT (Inner Circle Trader) framework. The idea is simple: one violent candle moves price fast and leaves a "vacuum" where almost nobody got filled. The theory says market makers rebalance and price tends to come back and fill at least half of that vacuum. So the bot waits for price to retrace into the gap, enters, takes profit at the 50% level (consequent encroachment in ICT-speak), and stops out outside the gap. Detection pipeline: Pre-filter: at least a 3-sigma move and 5x volume expansion Three-candle gap detection, scanning 1H and 4H It tracks the top 100 contracts around the clock, even when it holds nothing 2. The part nobody tells you: saying no Textbooks show you three candles and call it a day. Reality: a naive detector spits out dozens of signals a day and 99% of them are garbage. I built five gates to reject them. Each has a real log line from production: Freshness: gap older than ~100 candles? Drop it. [Freshness] SNXX 1H FVG 186 candles old > 24, drop ATR grade: gap width less than 0.5x ATR is a weak setup. [ATRGrade] width 0.16/ATR 0.37 = 0.43 < 0.5, weak C-grade Direction: don't long a coin that just pumped 14%, don't short one that dumped. [MoverDir] ETHFI 4H long rejected: +14.4% in 24h Depth: if the resting order is 6% off price, you're catching a falli

Last week a tweet went viral claiming that people complaining about LLM-generated bloat would "eat crow" once everything gets rewritten in hand-optimized assembly. Dan Luu, the engineer behind some of the most cited performance writing on the internet, responded with an essay titled "There's no reason for software to be slow anymore." It hit 620 points on Hacker News in about a day, and its argument should change how every Java team spends its next sprint. The core claim is simple and backed by real experiments: performance work that used to require a rare specialist can now be done by anyone who can type a few sentences. Luu quantifies it. The human-time cost of an optimization has dropped by what he calls "frequently 1000x / 10000x / 1000000x." He had an agent do workload-specific optimization of his own ripgrep usage, and launching it took about 2 minutes of his time. Jamie Brandon, a strong performance engineer, took Anthropic's public performance takehome exercise, then let Claude pick up where he left off. Claude got a much better result. Looking at the diff, Brandon said some of the agent's optimizations were things he had thought of but not gotten to, and others were, in his words, "just crazy shit that I would never try unless I was working on this for weeks." If you have spent six years writing Spring Boot services like I have, your reaction is probably the same as mine: interesting for regex engines, but what does this mean for the average enterprise Java service? The honest answer is that most of us will never need a custom JIT. But the underlying shift, that measuring and trying an optimization now costs minutes instead of days, applies directly to the slow endpoints every real codebase accumulates. This article is a practical workflow for turning an AI agent loose on a slow Spring Boot hot path without letting it ship garbage. Full disclosure up front: the numbers I cite from Luu's essay are his experiments, not mine. The workflow below is the one I no

The Performance Wall: Why React Isn't a Data Buffer If you’ve ever built a real-time application—a trading dashboard, a crypto ticker, or a live sensor monitor—you’ve likely hit the "React Performance Wall." You pipe your WebSocket messages directly into useState , and suddenly, your browser becomes a stuttering, unresponsive mess. The culprit is simple but often misunderstood: React is a UI library, not a data buffer. When you treat React state as the ultimate source of truth for every single byte of incoming data, you are essentially asking React to trigger a reconciliation cycle for every packet. If your backend is pushing data at 1,000Hz, you are trying to force 1,000 renders per second. Even the most optimized React app cannot handle that. You are blocking the main thread, tanking your frame rate, and leaving your users with a "lag machine." The "Death by a Thousand Cuts" Problem React’s reconciliation process is brilliant, but it is not built to trigger 1,000 times a second. Every setState call schedules a render. If you have a complex component tree, each render triggers diffing, lifecycle hooks, and DOM updates. When updates arrive faster than the browser can paint (typically 60Hz or 16.67ms per frame), you create a backlog of "long tasks." The browser’s main thread becomes so busy trying to keep up with the data stream that it ignores user interactions like clicks or scrolls. Your UI stops being a tool and starts being a bottleneck. The Architectural Shift: Decouple Ingestion from Rendering The fix isn't to optimize your components; it's to change your architecture. You need to stop letting React "know" about every single data point. At York.ie, we achieved a 40% boost in responsiveness by implementing a Dam Pattern . Instead of pushing packets directly into state, we treat the data flow like a dam: the water (data) flows in at high pressure, but we release it to the UI in controlled, manageable bursts. The Implementation Strategy Buffer Ingested Data: Use

Originally published on tamiz.pro . The proliferation of IoT devices, 5G networks, and real-time data processing demands has catalyzed a fundamental shift in computing paradigms: the move from centralized cloud infrastructure to distributed edge computing. This architectural evolution brings data processing and storage closer to the source of data generation, minimizing latency, conserving bandwidth, and enabling autonomous operations. However, distributing compute power across a vast, often heterogeneous network of edge nodes introduces significant complexities, particularly concerning middleware—the connective tissue enabling communication and data flow—and its inherent challenges around security and scalability. This deep-dive will explore the architectural implications of edge computing on middleware, focusing on the critical facets of security and scalability that define success or failure in this distributed landscape. Table of Contents 1. Understanding the Edge Computing Paradigm 2. The Role of Middleware in Edge Architectures 3. Middleware Security Challenges at the Edge 4. Strategies for Securing Edge Middleware 5. Scaling Middleware in Edge Environments 6. Architectural Patterns for Scalable Edge Middleware 7. Practical Considerations and Best Practices 8. Frequently Asked Questions 1. Understanding the Edge Computing Paradigm Edge computing extends the capabilities of cloud computing by bringing computation and data storage closer to the 'edge' of the network, where data is generated. This can range from industrial IoT devices, smart city sensors, retail points of sale, autonomous vehicles, and even user devices like smartphones. The primary motivations for this shift include: Reduced Latency: Processing data locally eliminates round trips to a central cloud, crucial for real-time applications like autonomous driving or industrial automation. Bandwidth Optimization: Only aggregated or pre-processed data needs to be sent to the cloud, significantly reducin

It happened during a quiet, solemn moment at a funeral. I felt the vibration in my pocket, and for a split second, I panicked. I had silenced my phone before entering, but I had accidentally toggled it back to normal mode while checking an email earlier that morning. In that room, the sound of a notification ping felt like a gunshot. The embarrassment was immediate and visceral. It was a clear signal that I needed a better way to manage my device's sound profile, a system that didn't rely on my flawed human memory. We live in an era of hyper-connectivity, yet our phones are surprisingly dumb when it comes to context awareness. I found myself constantly manually adjusting volume sliders. Meetings, gym sessions, prayer times, movie theaters—the list of places requiring silence is endless. Most existing solutions were either too heavy, requiring complex IFTTT integrations that lagged, or they were privacy-invasive, requiring constant cloud syncing. I wanted something that lived locally on my device, respected my data privacy, and didn't turn my phone into a brick by noon. The core problem wasn't just the silencing; it was the cognitive load of having to remember to revert those changes, which is how you end up missing important calls for the rest of the day. To build Muffle, I had to solve the geofencing puzzle. The temptation for any Android developer is to fire up a LocationRequest with high-accuracy settings and just poll the GPS coordinates. That is the fastest way to destroy battery life and get your app killed by the Android system's battery optimizations. Instead, I leaned into the GeofencingClient API. It is designed precisely for this use case: it lets the system handle the heavy lifting of location monitoring at the hardware level, rather than keeping the radio awake in my application process. I configured the GeofencingRequest using GEOFENCE_TRANSITION_ENTER and GEOFENCE_TRANSITION_EXIT triggers. The magic happens in the PendingIntent that gets fired when th

We break down the different autonomous driving levels, and why Level 5 is so hard to achieve.

The Quest Begins (The "Why") I still remember the first time I opened a legacy codebase and felt like I’d stepped into a dark dungeon without a torch. The file was a single 800‑line function called processData . Inside, variables bore names like tmp , x , flag , and comments that tried to explain every line: // TODO: refactor this mess function processData ( input ) { let r = []; // result array for ( let i = 0 ; i < input . length ; i ++ ) { // loop over items if ( input [ i ] > 10 ) { // if value greater than threshold let v = input [ i ] * 2 ; // double it if ( v % 2 === 0 ) { // if even r . push ( v ); // add to result } } } return r ; } I spent three hours tracing why a certain edge case produced an empty array, only to discover the comment “if value greater than threshold” was outdated—the threshold had changed to 12 in a later commit, but the comment never got updated. The code lied, the comments misled, and I felt like a hero who’d just swung at a shadow. That frustration sparked a question: What if we could write code so clear that comments became unnecessary? Not because we’re lazy, but because the code itself tells the story. The Revelation (The Insight) The treasure I uncovered wasn’t a new framework or a slick library—it was a mindset shift: make the code self‑documenting through intention‑revealing names and small, focused functions . When a variable, function, or class name reads like a sentence, the reader can infer what’s happening without a side note. Think of it like reading a well‑written novel. You don’t need footnotes to understand that “She opened the door and stepped into the rain” means she’s going outside. The same principle applies to code: if you name a function filterValuesAboveThreshold , the intent is obvious. Why does this matter? Because comments decay. They become outdated, they get ignored, and they add noise. Self‑explanatory code, on the other hand, stays accurate as long as the name stays accurate. It also forces you to think ab

Async-first culture is making teams faster and lonelier — but the real damage isn't loneliness. It's that people stopped owning decisions they were never in the room to make. Picture a mid-sized software company, somewhere between Series B and exhaustion, that has gone fully async. No standups. No sprint reviews that anyone actually attends. A bot joins every Zoom, spits out a bulleted summary, drops it into Confluence, and everyone reads it — or doesn't — in their own time. The calendar is clean. The focus time is protected. And six months later, a major architectural decision made in a Loom video that half the engineering team never watched has quietly become the source of a simmering, passive-aggressive standoff between backend and platform. Nobody was in the room. Nobody feels they own it. This is the efficiency trap, and it's subtler than its critics usually admit. The problem isn't that async communication is bad. Much of what passes for a meeting in most organizations is theater: 77% of workers attend meetings that end in a decision to schedule yet another meeting, and 62% regularly sit through meetings that didn't even state a goal in the invite. Cutting that noise is not just reasonable — it's overdue. The real issue is what happens when the pendulum swings past optimization into avoidance, and teams start treating the absence of synchronous contact as a metric of operational maturity. Presence Isn't the Point — Participation Is There's a distinction the async evangelism wave has systematically glossed over: the difference between being informed and being part of something . A well-structured meeting summary tells you what was decided. It does not tell you that the decision was almost different, that someone pushed back and was overruled, or that the VP's sudden hedge on a key tradeoff means the whole thing is about to be relitigated in three weeks. Those signals live in tone, in the slight pause before someone agrees, in who did and didn't speak. In face-t

Enterprise vibe coding: the governance framework for shipping AI-generated apps to production Published: August 22, 2026 Category: Enterprise · AI Deployments Reading time: 9 minutes Author: NEXUS AI Team Gartner forecasts that 40% of new enterprise production software will be built using vibe coding techniques by 2028. A 2026 scan of more than 1,400 live vibe-coded applications found that 65% already had a security issue, and 58% shipped with at least one critical vulnerability. Those two numbers describe the same industry moving in opposite directions at once: adoption is outrunning governance. This post covers what a governance framework for enterprise vibe coding actually looks like, the five controls it needs, and where most teams get it wrong. What is enterprise vibe coding? Enterprise vibe coding is the practice of using natural-language prompts to generate application code, then governing that code through mandatory review, access control, and audit before it reaches production, rather than letting it ship straight from a prompt to a live endpoint. The term (coined by Andrej Karpathy in early 2025) originally described a fast, low-friction way for one person to build a prototype. What "enterprise" adds is the governance layer prototyping was never built for: staging environments, encrypted secrets, role-based access, and a record of who approved what. That distinction matters because the adoption curve and the risk curve are not moving together. The governance gap, in three numbers 40% of new enterprise production software will be built using vibe coding techniques by 2028, according to Gartner's May 2025 report "Why Vibe Coding Needs to Be Taken Seriously," as reported by CIO Dive . 65% of vibe-coded production applications had a security issue, in a 2026 scan of more than 1,400 live apps by the API security firm Escape.tech, reported via a Cloud Security Alliance research note . 58% of those same applications shipped with at least one critical vulnerabilit

A Dutch data regulatory authority said that Uber has to pay 824.9 million euros for violating the GDPR.

Android Auto YouTube Limitations: Is It Worth Using In The Car?

Diario de un bot que opera con dinero real — Entrada #0: el origen Todo empezó con un tuit. Uno de esos que seguramente también has visto: una captura de una wallet, "$100 convertidos en $10.000 en 24 horas con este bot de trading", flechas verdes, emojis de cohetes, y un "sígueme para más". Debajo, cientos de likes y gente pidiendo el enlace. Mi primera reacción no fue "quiero eso". Fue "eso es mentira". Y no hace falta ser matemático para verlo. Un retorno del 10.000% en un día no es una estrategia — es un billete de lotería premiado que alguien presenta como si fuera un método repetible. Si de verdad tuvieras un sistema que multiplica tu dinero por cien cada 24 horas, no lo estarías publicando en X pidiendo likes. Lo estarías usando en silencio hasta comprar una isla. El que regala el mapa del tesoro es porque el tesoro no existe — el verdadero producto que se vende en esos tuits no es el bot: eres tú, tu like, tu follow, tu atención. Así que no le di like. Pero me quedé pensando. La pregunta que sí valía la pena Descartado el humo, quedaba una pregunta honesta debajo: despojado de la mentira del 10.000%, ¿hay algo real ahí? Porque los bots de trading existen. La automatización de estrategias es legítima. Los mercados operan 24/7 y un programa no duerme ni entra en pánico. La idea de fondo —dejar que un sistema ejecute una estrategia con disciplina, sin la emoción que arruina las decisiones humanas— no es una estafa. La estafa es el número. La estafa es prometer un retorno imposible para vender seguidores. Entonces me hice la pregunta que inició todo esto: ¿qué pasa si alguien escéptico construye un bot de trading de verdad, con expectativas sobrias, y documenta la verdad completa — incluida la parte donde todavía no sabe si funciona? Esa es la serie que estás empezando a leer. Lo que es, y lo que no es Para que no haya malentendidos, porque tú y yo ya sabemos cómo suele terminar este tipo de contenido: Esto no es un tutorial de "hazte rico". No voy a mostrarte u

I work in a 200-person Class II medical device company with two QA/RA folks, three embedded-hardware teams, a small supplier-quality group, and engineers who want automation — not more paperwork. We needed an eQMS that unifies design control (DHF), CAPA, supplier records, and traceability into something engineers will actually use. I evaluated several vendors and want to share a short, practical comparison for teams with that shape. A quick regulatory framing before the list Standards that mattered to us: ISO 13485 and 21 CFR 820, plus MDR-adjacent traceability requirements for components when applicable. A note on risk and regulation posture: don’t treat Article 50 (AI oversight posture for medtech tools) like an apocalypse — it sits in the general enforcement stack. Plan for reviewability and traceability for any AI-assisted workflows, but don’t let fear of hypothetical fines block good tooling decisions today. What I judged vendors on Fit for medtech design control / DHF workflows Traceability across requirements, risk, design, and CAPA Integrations — especially with CRM/supplier systems and engineering tools Operational fit for a 200-person org (configurability, onboarding burden) Capacity for process automation (automated CAPAs, connected workflow) The list — what each vendor practically means for a team like mine 1) Greenlight Guru Why I put it first: Greenlight Guru explicitly lists medical-device as its industry. For a Class II shop that needs DHF, design-control-native workflows, and templates tuned to medtech terminology, that industry focus matters. Fit: Strong for device-focused workflows and teams that want a medtech-first product experience rather than a generic QMS. When to pick: You want a purpose-built medtech experience, quicker ramp for QA/engineering, and vendor docs/templates that speak design control and DHF. 2) Qualio Public positioning: Qualio serves general industries including medical-device and pharma. Fit: Good option if you want flexibil

AI can produce an answer in seconds. The answer may be clear, plausible, and even correct. Yet something about it can still feel wrong. I do not think this discomfort comes only from hallucinations or poor model accuracy. Sometimes the real problem is simpler: The AI returned an output, but it did not return the work in a form that another person can safely continue. This is not a new problem created by AI. It is the same problem we already have when delegating work to another person. What do we expect when we delegate work? Imagine a manager asking a team member: Please prepare a proposal for reducing next month's operating costs. The team member reviews several documents, compares multiple options, and replies: We should choose Option A. The requested conclusion has been delivered. But has the work really been handed back? The manager still does not know: What objective the team member optimized for Which documents and facts were examined Which assumptions and constraints were used Which alternatives were compared Why Option A was preferred Which conditions remain unverified What must be reconsidered if the situation changes The original request may not have explicitly demanded all of this. Even so, we normally expect a competent team member to understand the purpose of the assignment and to return enough information for someone else to review, approve, revise, and continue the work. That information is not additional reporting attached to the work. It is part of the handoff condition that makes delegation possible. AI often returns the conclusion without the handoff Now replace the team member with an AI assistant. The AI immediately recommends Option A and produces a polished explanation. Because the answer arrives so quickly and looks complete, it is easy to confuse the existence of an output with the completion of the work. But the same questions remain: How did the AI interpret the objective? What was considered in scope and out of scope? Which sources were a

Pydantic AI gives you a clean, typed agent: define an Agent , hand it tools, call agent.run(...) , and it loops — model call, tool call, model call — until it produces a validated result. The typed ergonomics are great. What the quickstart doesn't spell out is what the model receives on each pass of that loop. I read the run graph ( pydantic_ai_slim/pydantic_ai/_agent_graph.py on main ) to find out. The mechanism is structural, and it's the same shape I found in the OpenAI Agents SDK and smolagents. One list, appended twice per turn Each run holds a single mutable conversation list on its state: message_history : list [ _messages . ModelMessage ] = dataclasses . field ( default_factory = list [ _messages . ModelMessage ]) On every model step the graph appends to it — first the outgoing request, then the model's response: ctx . state . message_history . append ( self . request ) ... ctx . state . message_history . append ( response ) Nothing is removed. The list only grows: request, response, request, response — with tool calls and, crucially, tool outputs riding inside those messages. The full list is re-sent every step When the graph builds the input for the next model call, it takes the entire accumulated history — a full copy: messages = ctx . state . message_history [:] ... messages [:] = _clean_message_history ( ctx . state . message_history ) That [:] is the whole conversation to date. So on step 1 the model sees your prompt; on step 2 it sees your prompt + step 1's request + step 1's response (including the tool output); on step 5 it sees all of that plus steps 2–4. The payload you pay for grows every single step, and the heaviest passengers are usually the tool outputs — the search results, file contents, and API responses you least want re-uploaded five times. Why it's quadratic, and why nothing warns you A run of n steps sends roughly 1 + 2 + 3 + … + n copies of history — O(n²) cumulative tokens in the step count. A 3-step agent is fine. A 12-step agent th

자동으로 돌아가게 만든 것과, 자동으로 끝까지 처리되는 것은 다른 문장이었습니다 이 시스템은 사람 승인 없이 스스로 판단하고 매매하는 걸 목표로 설계했습니다. 실계좌 주문 실행과 안전장치 (새 창)도 그 목표에 맞춰 만들었습니다. 그런데 최근 한 달 사이 실계좌에서 세 번, 사람이 직접 개입해야 하는 상황이 있었습니다. 세 사례 모두 "왜 자동 로직이 이 상황을 못 넘겼는지"의 구조가 서로 달랐습니다. 1. 배분 규칙이 특정 주문을 구조적으로 굶겼다 특정 종목 하나가 여러 날째 매도 계획이 서 있는데도 계속 팔리지 않는 걸 발견했습니다. 시스템은 매일 이 종목을 매도 후보로 올렸지만, 실제 주문까지는 못 갔습니다. 원인은 하루 매매 한도를 여러 라운드에 나눠 배분하는 규칙이었습니다. 이 종목의 주문 금액이 그날 남은 매도 한도보다 항상 컸습니다. 라운드 순서를 아무리 바꿔도 통과할 수 없는 구조였습니다. 한도 자체는 정상 작동하고 있었습니다. 문제는 "이번엔 못 나가도 다음 기회에 나간다"는 전제가 이 종목엔 애초에 성립하지 않았다는 점입니다. 잔여 한도가 매번 주문 금액보다 작으면, 기회는 계속 오지만 한 번도 충분하지 않습니다. 당장 못 나간 주문 1건은 사람이 직접 처리했습니다. 실계좌에서 이뤄진 되돌릴 수 없는 매도였습니다. 이후 배분 규칙 자체를 손봐서 같은 구조로 다시 굶는 일이 없도록 정리했습니다. 2. 안전장치가 스냅샷과 누적치를 혼동했다 다른 날엔 반대 방향의 사고가 있었습니다. 누적 손실을 감지하는 안전장치가 정상적인 매수 2건을 잘못 차단했습니다. 지수는 그날 거의 보합이었는데, 이 안전장치가 재는 손실률은 훨씬 크게 찍혀 있었습니다. 원인을 보니 이 장치는 "고점 대비 누적 하락"을 감지하는 용도였는데, 정작 비교하는 현재값은 장중 순간 스냅샷이었습니다. 장중 잠깐의 변동이 누적 지표를 밀어 올려서, 실제로는 발동하면 안 될 상황에서 발동한 겁니다. 누적을 재는 장치와 순간을 재는 장치가 뒤섞여 있었던 셈입니다. 막힌 매수 2건은 사람이 판단해서 직접 집행했습니다. 이후 이 안전장치가 장중 순간값이 아니라 "그날 마감 대 전날 마감" 기준으로만 반응하도록 구조를 바꿨습니다. 장중 급락에는 이제 다른 안전장치가 대신 반응하도록 역할을 나눴습니다. 3. 개입 경로 자체가 "새로 사는 경우"를 몰랐다 두 번째 사례를 수습하는 과정에서 사고가 하나 더 있었습니다. 수동으로 낸 주문을 원장에 반영하는 도구를 썼는데, 반영이 안 되고 조용히 빠졌습니다. 이 도구는 사람이 손으로 낸 거래를 세 가지 경우 중 하나로 분류합니다. 기존 보유 종목을 판 경우, 기존 보유 종목을 더 산 경우, 그리고 시스템과 무관한 거래인 경우입니다. 그런데 이번 매수는 원장에 없던 새 종목을 사람이 처음 사들인 경우였습니다. 세 분류 중 어디에도 안 맞았고, 도구는 이걸 "시스템과 무관한 거래"로 잘못 넘겼습니다. 그 결과 실제로는 산 자산이 잠깐 원장 밖에 있는 것처럼 표시됐습니다. 이 도구는 애초에 사람 개입을 위해 만든 경로였습니다. 그런데 그 경로를 설계할 때, "사람이 아예 새로운 자리에 처음 진입하는 경우"는 상정하지 않았습니다. 개입 경로 자체가 개입의 한 형태를 놓치고 있었던 셈입니다. 순서(먼저 다른 매도를 부기하고, 그다음 이 매수를 부기)를 지켜서 바로 수습했고, 검증 결과 원장과 실계좌 잔고는 정확히 일치했습니다. 분류 로직에 이 경우를 추가하는 건 아직 남은 과제입니다. 세 사례를 묶어보면 셋 다 "자동으로 처리되게 만들었다"와 "실제로 끝까지 처리된다"가 다른 문장이라는 걸 보여줬습니다. 첫 번째는 규칙이 있었지만 그 규칙이 특정 입력에서 절대 통과할 수 없는 구조였습니다. 두 번째는 장치가 있었지만 재는 대상(순간 대 누적)이 설계 의도와 어긋나 있었습니다. 세 번째는 사람 개입을 위한 경로가 있었지만 그 경로 자체가 특정 개입 형태를 몰랐습니다. 세 가지 모두 "자동화가 이 케이스를 놓칠 수 있다"는 걸 사전에 안 게 아니라, 실제로 놓친 뒤에야 알았습니다. 일반화하면 완전자동을 목표로 설계할수록,