标签：#war

The biggest losses in software history were, with one deliberate exception, not attacks. They were silent, correlated, self-inflicted — and they teach the exact risk autonomous AI agents are about to make expensive again. At 9:30 in the morning on August 1, 2012, Knight Capital Group was one of the largest trading firms in the United States, executing a sixth of all the volume on the New York Stock Exchange. By 10:15 it was, for practical purposes, finished. In those forty-five minutes a piece of its own trading software (not a hacker's, its own) fired more than four million unwanted orders into the market, accumulating roughly $7 billion in positions the firm never meant to hold and a loss of about $440 million by the time humans understood what their machine was doing. The cause, documented in the SEC's administrative proceeding, was almost insultingly small: a deployment that updated seven of eight servers. The eighth still carried a dormant piece of code called Power Peg, retired years earlier, and the new release reused the old feature flag that woke it up. No one attacked Knight Capital. The market data was accurate, the exchange functioned perfectly, and every system reported itself healthy while the company bled ten million dollars a minute. That shape (no adversary, no alarm, one change propagating everywhere at once) turns out to be the shape of almost every entry on the list below. We've written before about the biggest bug-bounty payouts in history , the ledger of what it costs when someone does attack. This is the other ledger, the bigger one: what software has cost when nobody attacked at all. Every figure below states what it counts, and comes from a primary or authoritative source (inquiry boards, SEC filings, statutory inquiries) linked at the end. The ledger 1. CrowdStrike outage (2024) — roughly $5.4 billion in direct losses to Fortune 500 companies alone (estimate). One faulty content update to the Falcon Sensor security agent blue-screened Windo

I am interested in agentic coding for the same reason I care about good engineering process in general: I want work to move forward in a way that is inspectable, repeatable, and resilient once the task gets messy. A lot of AI-assisted coding still feels like improvisation. You ask for something, get a result, adjust the prompt, try again, and hope the useful reasoning is still somewhere in the scrollback. That can work for tiny edits. It gets much less convincing when the task starts touching architecture, tests, review, or pull requests. What I want instead is a workflow where the model helps me think and execute, but inside a structure I can inspect afterwards. I want artifacts, gates, and something I can resume tomorrow without reconstructing the entire mental state from memory. That is why I use po8rewq/agentic-skills . It gives me a practical way to do agentic coding as an engineering workflow rather than as a long sequence of chat turns. A task moves through requirements, architecture, implementation, checks, review, and pull request creation. Each stage leaves something I can read, verify, and challenge. What makes this interesting to me The interesting part is not just that there is a CLI. Plenty of tools have a CLI. What matters to me is that it turns AI-assisted coding into a staged system: requirements force the task to become explicit architecture makes risks visible before code is written implementation happens against a plan instead of against a vague prompt checks and review happen as part of the flow, not as an afterthought runs are resumable, so interruptions do not destroy context That changes the feel of the work quite a bit. Instead of asking "what should I prompt next?", I am usually asking "what stage is this task in, and what should exist before I move on?" Where this really clicked for me was when I noticed I was spending less energy trying to preserve context in my head and more energy evaluating actual outputs. What the repository actually

The call-order change came back pass-with-risk. I read the recommendation, saw it had a name and a reason, and felt the task close. Then I looked at the row under it. How was this verified: not run. Nobody had run the queue. I had a label. I did not have proof. This is Part 6 of The Contract Think produced a brief. Plan produced a gate. Build executed inside it. Review scored every requirement against a verdict instead of an impression. Review reads the diff and the plan and decides whether one satisfies the other. It does not run the queue. It cannot. Its whole job is judgment about what the code should do. Test is where someone finally checks what the code actually does. I had been treating those two as the same step. They are not. Test asks one question, and a verdict is not the answer For every active requirement, Test asks how it was verified. Command run, manual QA, or a comparison against known-good output. One of those three, or a written reason none of them ran. Not a recommendation. Not a risk level. Evidence. I built the matrix against the plan's requirements and filled in each row. Most had a command behind them. The call-order requirement had nothing. The cell read not run, and it sat directly below a pass-with-risk that already carried a name and a reason. That name had almost been enough for me. A named risk feels handled. It is not. It is a risk with a label on it, waiting for someone to actually look. So I ran the queue Three notifications, all with a real reason to fire within the same tick. The scheduler picked them up and ordered them by priority instead of arrival. Two landed in the sequence the requirement wanted. The third jumped ahead of a lower-priority notification that was still mid-processing. The change worked almost every time. Under one timing condition, it did not. That is the gap a verdict cannot see. Review had marked the requirement partial because the wording left the mechanism open. Running it found a real failure inside the mech

US military drone losses in Iran war spur Pentagon call for cheap replacements.

Apple has signed a $30 billion+ multiyear deal with Broadcom to design and produce more than 15 billion U.S.-made custom wireless connectivity chips for Apple products.

With funding from tennis star and angel investor Venus Williams, the French app WeWard says that it increases walking time by almost 25%.

Enterprises are now talking about AI coding agents in a very predictable way. The first question is usually: "How much more code can it help us generate?" It is not a wrong question. But if that is the only question, the ROI calculation will probably be wrong. Because enterprises are not really buying "more code." They are buying: faster delivery less rework lower maintenance cost better developer experience more stable software quality more controllable security and compliance risk faster translation from product capability to business value Code generation is an input. It is not the outcome. That distinction matters. An AI coding agent can help developers write functions, fix bugs, add tests, generate documentation, understand codebases, and refactor legacy systems. That sounds powerful. But the enterprise question is not: "How many lines of code did it generate today?" The better question is: Did that code reach production faster? Did incidents go down? Did the team spend less time on repetitive work? Did customers get value sooner? If the answer is unclear, generating 100,000 lines of code a day may simply mean producing technical debt faster. The short version: AI coding agent ROI does not end inside the IDE Many teams start measuring AI coding tools with the most obvious numbers: code suggestion acceptance rate lines of code generated number of active users number of prompts time saved on individual tasks These metrics are useful. But they mostly show that the tool is being used. They do not prove that the enterprise is getting value. Enterprise ROI has to be measured across software delivery, quality, risk, and business outcomes. In other words, an AI coding agent is not just a point solution for individual efficiency. It affects the entire software value stream: Request -> Design -> Coding -> Review -> Testing -> Deployment -> Monitoring -> Feedback -> Business outcome If you calculate value only inside the "coding" box, you miss the bigger picture. Why "amo

A cybersecurity startup dangling millions of dollars to acquire zero-day security vulnerabilities in popular software is run by a pair of far-right conspiracy theorists and convicted felons whose most recent ventures included fake intelligence companies and a now-defunct AI-based lobbying platform they operated under assumed names.

A few years ago, securely accessing my servers at home or in a small office from outside was always a headache. Dynamic IP addresses, complex port forwarding settings in the modem interface, and even the need to open rules in the network firewall always presented me with a new problem. Especially outside of corporate projects, when setting up my side products or test environments, these processes caused a loss of time. Cloudflare Tunnel offered a game-changing solution for such self-hosted services. Essentially, a cloudflared daemon (tunnel client) on your local network establishes a single outbound connection to Cloudflare, eliminating the need to open any firewall holes for incoming connections. This model provides significant advantages over traditional methods in terms of both security and ease of use. Why Were Traditional Self-Hosting Methods Challenging? Exposing a self-hosted service to the internet typically brings multiple technical challenges. First, most home or office internet connections have dynamic IP addresses, meaning your IP address changes from time to time. In this case, to access your service via a fixed domain name, you need to resort to solutions like Dynamic DNS (DDNS), which adds an extra dependency and sometimes causes delays. Second, and perhaps most importantly, is the necessity of opening port forwarding rules in your network firewalls (modem/router firewall). This means you need to direct specific ports to your internal IP address to allow incoming connections. This process both increases security risks (internet-facing ports invite brute-force attacks) and causes access problems if not configured correctly. I once experienced a serious panic during a client project when I accidentally exposed a critical internal service port to the outside while dealing with these ports. Since that day, I approach such manual interventions with more skepticism. ℹ️ Port Forwarding Risks Every port you open in your network firewall creates a potential at

When analyzing supply chain data in a production ERP, the idea of sending critical information to a publicly accessible cloud-based LLM always bothered me. Processing sensitive data like proprietary production plans, customer lists, or financial details on third-party servers was an unacceptable risk for me. In such situations, the only way to ensure data security is to run the LLM in a local environment, under our own control. This is where Ollama comes in. Ollama is a tool that allows you to easily run large language models (LLMs) on your local system. This way, you can set up your own AI assistant without an internet connection and without the risk of leaking your data, securely automating your sensitive business processes. In this post, I will step-by-step explore why Ollama is important, how to install and use it, and the data security advantages it offers. Why Ollama is Important: Does it Provide Data Security and Control? The potential of LLMs in modern software development processes and operational workflows is huge, but the privacy and security of corporate data are always a primary concern. Especially in high-security environments like a bank's internal platform or in the financial calculators of my own side product, I need to process user data without sending it externally. In these scenarios, cloud-based LLM services often create a new risk factor rather than a solution. Ollama is designed to close this critical data security gap. By running models on your local machine, it prevents your data from leaving your company network or personal computer. This is a vital advantage, especially for those working in sectors subject to data protection regulations like GDPR and KVKK. Furthermore, because it doesn't require an internet connection, you can continue to benefit from LLM capabilities even in offline environments or during network outages. Since you have full control, you can manage from start to finish which model runs with which data, how much resource t

Software architecture debates have a problem that most other engineering disciplines don't: the alternative was never built. When a bridge fails, the failure is physical, attributable, and measurable against every other bridge that didn't. The engineering decisions that caused it can be isolated, traced, and corrected — not just in theory, but in the next bridge, because the material itself produces feedback that no amount of professional opinion can override. Steel deflects. Concrete cracks. Physics doesn't care what the architect believed. Software produces no equivalent feedback. A system built around the wrong abstractions compiles, runs, ships, and passes its tests just as readily as one built around the right ones. A bug introduced by a misaligned domain model looks identical, from the outside, to a bug introduced by a typo. A feature that took three times longer than it should have, because the structure made it harder than the business logic warranted, produces no artifact that distinguishes it from a feature that was simply difficult. The cost is real. The cause is invisible. This is the unfalsifiability problem, and it runs deeper than "we can't measure everything." It means that when a system becomes expensive to change, the diagnosis almost always lands on the wrong variable. The domain is complex. The requirements changed. The previous team was careless. Almost never: the structure was wrong, and the structure was wrong because nobody ever built the other version of it to compare against. That version doesn't exist, it never will, and every architectural argument in the industry is conducted in its absence. This would be a purely philosophical problem if there were nothing to do about it. There is something to do about it — but it requires accepting that the standard metric for software quality, whether it works, is measuring the wrong thing entirely. The Metric That Hides the Problem The natural substitute for "is this good engineering" is "does it wor

Why HDI PCB Manufacturing Starts Long Before the First Hole Is Drilled When people think about PCB manufacturing, they usually imagine drilling, plating, imaging, etching, solder mask, and surface finishing. For conventional PCBs, that assumption isn't too far from reality. For HDI (High Density Interconnect) PCBs, however, manufacturing actually begins long before any physical production starts. The success of an HDI project is often determined during engineering review rather than on the factory floor. Manufacturing Starts with Design Decisions A PCB layout may pass every design rule check inside CAD software while still being difficult to manufacture efficiently. Typical examples include: unnecessary stacked microvias excessive sequential lamination extremely aggressive trace and space dimensions unrealistic copper balancing inefficient stack-up planning None of these issues are fabrication defects. They are engineering decisions. The earlier they are identified, the lower the overall project cost becomes. The Stack-Up Is More Important Than Many Engineers Expect One of the biggest misconceptions is that increasing the layer count automatically solves routing problems. In reality, a carefully planned stack-up usually provides greater benefits than simply adding more copper layers. A good stack-up improves: signal integrity impedance consistency EMI performance power distribution thermal behavior More importantly, it creates a PCB that is easier to manufacture repeatedly with stable quality. HDI Is a Balance Between Performance and Manufacturability Many first-time HDI designs focus only on routing density. Experienced engineers usually focus on manufacturability. For example: Should this microvia really be stacked? Can staggered vias achieve the same result? Is another lamination cycle actually necessary? Can the BGA fan-out be optimized differently? Each decision influences fabrication complexity, yield, lead time, and production cost. Why DFM Matters More for H

Reading Lilian Weng's harness engineering survey as a reliability engineer — what self-improving harness papers actually show, and the three invariants every working loop converges on.

Most web applications contain at least one vulnerability from the OWASP Top 10. A typical security audit takes 2-3 weeks and costs upward of $10,000. An LLM can compress the initial audit down to a few hours because it scans code for patterns rather than specific CVEs. Below are 15 vulnerabilities found while auditing production code with Claude. Each includes the vulnerable code, the fixed version, and a prompt to reproduce the finding. Classification follows OWASP Top 10 (2021). Order reflects frequency of occurrence: most common first. Methodology: how to run an AI security audit The audit consists of three passes. First, a broad scan: the LLM receives the entire project and looks for vulnerability patterns. Second, deep analysis: each identified pattern is verified in context (middleware, ORM, framework). Third, verification: manual review of every finding, because LLMs produce false positives. Prompt for the broad scan: Perform a security audit of this code. For each finding, include: 1. CWE ID and name 2. OWASP Top 10 category 3. Severity (Critical/High/Medium/Low) 4. The vulnerable code snippet 5. Attack vector -- exactly how an attacker would exploit this 6. Fixed code Ignore stylistic comments. Focus on security only. Start with injection attacks, then broken access control, then the rest. This prompt works because it defines the output structure and prioritizes categories. Without explicit instructions, the LLM mixes critical vulnerabilities with remarks about email validation. More on structured AI code review: AI Code Review Checklist . A03:2021 -- Injection 1. SQL Injection via string concatenation The most common finding. Shows up even in projects using an ORM, because developers switch to raw queries for complex filters. Vulnerable code: // API endpoint for user search app . get ( ' /api/users ' , async ( req , res ) => { const { search , sortBy } = req . query ; const query = ` SELECT id, name, email FROM users WHERE name LIKE '% ${ search } %' ORDER

A lot of AI memory systems start with the same temptation: "Just save the useful thing." That sounds harmless until the knowledge base becomes a junk drawer. Half the notes are too specific, a few are duplicates, some are obsolete, and nobody knows which ones the agent should trust. In ai-assistant-dot-files , the memory system is deliberately slower. It uses a promotion lifecycle: Capture -> Candidate -> Audit -> Approve -> Index -> Retrieve -> Expire That lifecycle is documented in docs/runbooks/memory-engineering.md , and the important word is not "capture." It is "candidate." Nothing writes directly to memory The framework has a durable memory layer: Knowledge Items in shared/knowledge/ , ADRs in docs/adrs/ , the domain dictionary, team topology, a feature archive, and a registry at shared/memory-registry.json . But a lesson from a delivery does not jump straight into shared/knowledge/ . It first becomes a Candidate Record. That record has required fields: Source Type Evidence Tags Expiration condition Then memory-engineer audits it: Is it reusable? Is it already covered? Is it too speculative? Does it belong as a Knowledge Item, or should it become a rule change, prompt edit, or ADR instead? Only after that does a human approve the destination. The design is intentionally similar to code review. Durable memory changes future behavior, so they deserve a paper trail. Rejection is a feature One of my favorite parts of the memory runbook is that it has explicit rejection rules. Do not promote a memory when it is: a one-off already covered too speculative That makes "zero candidates promoted this cycle" a healthy result, not a failure. This is where memory engineering starts to look less like note-taking and more like gardening. The point is not to preserve every leaf. The point is to keep the soil useful. Expiration matters The lifecycle also includes expiration. A Knowledge Item can become stale when the underlying code, agent, or pattern changes. It can be supers

As much as 50 percent of some teams affected by reductions, and more could be coming.

Async route: vì sao Promise reject trong Express 4 không tới error middleware, và cách Fastify tự bắt lỗi Một async route handler trông giống một handler bình thường, nhưng cơ chế truyền lỗi của nó khác hẳn. Express 4 bọc lời gọi handler bằng try/catch đồng bộ; một hàm async return về một Promise trước khi Promise đó settle, nên try/catch đồng bộ không bắt được rejection. Kết quả: lỗi rơi ra khỏi handler dưới dạng unhandledRejection , không tới error middleware, không được gửi thành 500 JSON , và từ Node 15 mặc định làm process crash. Fastify đi hướng khác — pipeline hook tự await handler, rejection được framework bắt và chuyển tới setErrorHandler . Hiểu điểm khác biệt này giải thích vì sao cùng một dòng code await db.query(...) lại có hai hệ quả production hoàn toàn khác nhau giữa hai framework. Cơ chế hoạt động Trong Express 4, Layer.handle_request gọi handler đồng bộ trong một khối try/catch . Nếu handler ném exception đồng bộ, catch được, forward qua next(err) . Nhưng handler async không ném — nó return về một Promise. Đến khi Promise reject, control đã ra khỏi khối try/catch từ lâu: // Đây là bản giản lược của cách Express 4 gọi handler: try { const ret = handler ( req , res , next ) // với async handler, ret là Promise (đã pending) // handler đã "return" — try/catch không còn hiệu lực với reject xảy ra sau } catch ( err ) { next ( err ) // chỉ chạy khi handler ném đồng bộ } Vì Express 4 không await ret và cũng không ret.catch(next) , một Promise reject không có bên nào bắt. Node emit unhandledRejection ở tầng process. Cách sửa chuẩn là wrapper: // asyncHandler: bọc handler async, forward reject vào next(err) export const asyncHandler = ( fn ) => ( req , res , next ) => Promise . resolve ( fn ( req , res , next )). catch ( next ) app . get ( ' /orders/:id ' , asyncHandler ( async ( req , res ) => { const order = await db . orders . findById ( req . params . id ) if ( ! order ) { res . status ( 404 ). json ({ error : ' not_found ' }); return } res . json ( order

Google's upcoming event in August will introduce new Pixel devices.

Ars analysis suggests the 9-year-old console could keep selling for years.

I hesitated to write this. Not because I don’t have an opinion about AI in software engineering, but...