<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/"><channel><title>Will Hackett (English (US))</title><description>I’m Will, co-founder and CTO of flowstate. Mostly software, AI and economics. Occasionally a minor inconvenience becomes an essay with sources.</description><link>https://willhackett.com/</link><language>en-US</language><managingEditor>hello@willhackett.com (Will Hackett)</managingEditor><webMaster>hello@willhackett.com (Will Hackett)</webMaster><atom:link href="https://willhackett.com/en-us/feed.xml" rel="self" type="application/rss+xml"/><item><title>The impossibility of measuring AI productivity</title><link>https://willhackett.com/en-us/residue/</link><guid isPermaLink="true">https://willhackett.com/en-us/residue/</guid><description>I wanted a way to tell whether AI was making work better. I ended up arguing with ticket
counts, cost accounting and my own assumptions. This is the method I would test, and the
long list of reasons to be careful with it.
</description><pubDate>Wed, 07 Oct 2026 02:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>ai</category><category>business</category><category>finance</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Never store data you don&apos;t need</title><link>https://willhackett.com/en-us/ask-later/</link><guid isPermaLink="true">https://willhackett.com/en-us/ask-later/</guid><description>A website needs to know whether I&apos;m old enough to buy wine. It doesn&apos;t need a copy of my passport. Companies keep collecting far more than they need, then act surprised when someone steals it. Here&apos;s how to build it better, and what to do if your data&apos;s already out there.
</description><pubDate>Mon, 28 Sep 2026 05:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>security</category><category>opinion</category><category>technical</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Two agents I use to stay focused</title><link>https://willhackett.com/en-us/follow-through/</link><guid isPermaLink="true">https://willhackett.com/en-us/follow-through/</guid><description>Two scheduled agents turn my meetings, inbox and calendar into one daily note that keeps me focused and knows who I owe what. Here&apos;s the setup and every prompt.</description><pubDate>Fri, 25 Sep 2026 11:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>ai</category><category>productivity</category><category>personal</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Apple&apos;s monopoly on coherence</title><link>https://willhackett.com/en-us/nowhere-to-go/</link><guid isPermaLink="true">https://willhackett.com/en-us/nowhere-to-go/</guid><description>Downloaded. Ready. Connected. Apparently. Apple built the only ecosystem that feels like one computer. Lately that computer seems unsure what it&apos;s doing, and there is still nowhere else to go.</description><pubDate>Wed, 02 Sep 2026 07:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>opinion</category><category>personal</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Buy tokens, rent GPUs, or own the rack?</title><link>https://willhackett.com/en-us/son-of-anton/</link><guid isPermaLink="true">https://willhackett.com/en-us/son-of-anton/</guid><description>Open-weight models are now good enough to make self-hosting a financial question rather than a capability one. I modeled four options, an office rack, colocation, rented GPUs and per-token APIs, to find where each one actually wins. The answer turns on utilization, and it isn&apos;t the one the internet keeps giving.
</description><pubDate>Thu, 30 Jul 2026 16:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>ai</category><category>infrastructure</category><category>finance</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>The Frontier</title><link>https://willhackett.com/en-us/frontier/</link><guid isPermaLink="true">https://willhackett.com/en-us/frontier/</guid><description>The capability gap between closed and open AI models, collapsing. Real LMArena data, told as a city.</description><pubDate>Tue, 23 Jun 2026 08:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Every worktree gets its own database</title><link>https://willhackett.com/en-us/wtdb/</link><guid isPermaLink="true">https://willhackett.com/en-us/wtdb/</guid><description>Git worktrees only clone your code, so every worktree uses the same local database. One branch&apos;s migration breaks the rest. A schema change breaks the APIs. wtdb is a small Go tool that gives each worktree its own cloned Postgres database, env file, and node_modules.
</description><pubDate>Sun, 14 Jun 2026 14:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>git</category><category>programming</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>US Gov just killed Fable (for now)</title><link>https://willhackett.com/en-us/pentagon-party/</link><guid isPermaLink="true">https://willhackett.com/en-us/pentagon-party/</guid><description>The US government just ordered Anthropic to switch off two of its best models for everyone, citing
national security over a jailbreak the company disputes.
</description><pubDate>Sat, 13 Jun 2026 01:40:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>ai</category><category>opinion</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Fable 5 isn&apos;t a new Opus, it&apos;s a new ceiling</title><link>https://willhackett.com/en-us/fable-5-new-ceiling/</link><guid isPermaLink="true">https://willhackett.com/en-us/fable-5-new-ceiling/</guid><description>Anthropic shipped Claude Fable 5 at $10/$50 per million tokens, a full tier above Opus and
exactly double its price. It&apos;s a flagship, not a demonstrator, and it&apos;s the cleanest proof yet
that AI pricing went K-shaped: the floor under Opus was set so the ceiling could be raised.
</description><pubDate>Tue, 09 Jun 2026 22:50:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>ai</category><category>economics</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>A classifier is mostly plumbing</title><link>https://willhackett.com/en-us/classifier-hell/</link><guid isPermaLink="true">https://willhackett.com/en-us/classifier-hell/</guid><description>Your team is paying flagship AI prices for everyday tasks. Here&apos;s how routing and attribution turn a $3M undifferentiated OpEx bill into an amortizable asset.</description><pubDate>Tue, 09 Jun 2026 07:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><category>ai</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Locking the new Siri out of the EU market is strange</title><link>https://willhackett.com/en-us/siri-eu-lockout/</link><guid isPermaLink="true">https://willhackett.com/en-us/siri-eu-lockout/</guid><description>Apple introduced Siri AI for iOS 27 but not in the EU, blaming the Digital Markets Act, while
giving EU users the same assistant on the Mac. The line tracks which platforms the law
designates, not any real difference in risk.
</description><pubDate>Mon, 08 Jun 2026 23:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>opinion</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Turn a $3m AI bill into $1.9m</title><link>https://willhackett.com/en-us/opus-tax/</link><guid isPermaLink="true">https://willhackett.com/en-us/opus-tax/</guid><description>Most teams overpay for AI twice without noticing: they run everything through the default flagship model, and they buy the tokens on a metered plan they can&apos;t see into. An intelligent proxy fixes both. Here is the math, a calculator you can drag, and where Flowstate fits.
</description><pubDate>Mon, 08 Jun 2026 14:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>ai</category><category>business</category><category>finance</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Slightly coherent rant about a menu</title><link>https://willhackett.com/en-us/leave-it-alone/</link><guid isPermaLink="true">https://willhackett.com/en-us/leave-it-alone/</guid><description>Microsoft moved the menu again. A fast, funny tour of why integration beats flexibility, and
why a company that stares at its rival instead of its user keeps losing the one person who
matters.
</description><pubDate>Fri, 05 Jun 2026 07:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>opinion</category><category>personal</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>On writing</title><link>https://willhackett.com/en-us/on-writing/</link><guid isPermaLink="true">https://willhackett.com/en-us/on-writing/</guid><description>How posts on this site actually get made, from a thought to a published page, and the per-sentence commenting layer I use to chase down where a draft is breaking. The comment box is open on this one. Try it.
</description><pubDate>Mon, 01 Jun 2026 08:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>writing</category><category>meta</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Observations on AI agent token consumption</title><link>https://willhackett.com/en-us/agent-token-consumption/</link><guid isPermaLink="true">https://willhackett.com/en-us/agent-token-consumption/</guid><description>A new paper from Stanford, Michigan, DeepMind and All Hands is the first open empirical study of how AI agents actually spend tokens at scale. The findings line up closely with what we are seeing at Flowstate, from a different angle.
</description><pubDate>Sat, 16 May 2026 08:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>ai</category><category>economics</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>How do you tell who&apos;s thinking?</title><link>https://willhackett.com/en-us/borrowed-cognition/</link><guid isPermaLink="true">https://willhackett.com/en-us/borrowed-cognition/</guid><description>AI is fantastic for extending thinking. The trouble is that the people crossing the line from extension to substitution cannot tell, from the inside, that they have. Including, possibly, the person who wrote this.
</description><pubDate>Thu, 07 May 2026 17:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>ai</category><category>opinion</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Peter Whitehouse – teaching over measurement</title><link>https://willhackett.com/en-us/peter-whitehouse/</link><guid isPermaLink="true">https://willhackett.com/en-us/peter-whitehouse/</guid><description>A tribute to Peter Whitehouse, my high-school IT teacher in Brisbane,
who taught me memory management at fifteen, ran the IT club out of his
spare time, and quietly built the architecture-thinking I lean on every
day. Retired in 2022. Needed now more than ever.
</description><pubDate>Wed, 29 Apr 2026 21:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>career</category><category>opinion</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Some things never change</title><link>https://willhackett.com/en-us/some-things-never-change/</link><guid isPermaLink="true">https://willhackett.com/en-us/some-things-never-change/</guid><description>The browser wars settled into a three-horse race. The AI race looks suspiciously similar: three players, scrappy upstarts, and one kid in the corner sucking on a jar of glue.
</description><pubDate>Sun, 26 Apr 2026 21:10:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>ai</category><category>opinion</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>My advice to a younger me</title><link>https://willhackett.com/en-us/lessons/</link><guid isPermaLink="true">https://willhackett.com/en-us/lessons/</guid><description>In 2012 Scott Weiss wrote that the best preparation for becoming a founder is working at a startup. Fourteen years and several failures later, here are the seven lessons I wish I&apos;d actually listened to, on people, product, ownership, failure, and knowing when to step away from the keyboard.</description><pubDate>Wed, 15 Apr 2026 13:50:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>startups</category><category>career</category><category>leadership</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>12 seconds to scan 3,000 rows, oops</title><link>https://willhackett.com/en-us/enable-nestloop/</link><guid isPermaLink="true">https://willhackett.com/en-us/enable-nestloop/</guid><description>Postgres&apos; WindowAgg node doesn&apos;t propagate cardinality estimates. That single bad estimate can poison every downstream join decision. Sometimes the pragmatic fix is the right one.
</description><pubDate>Tue, 14 Apr 2026 11:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Workforce Engineering has always existed. We just never named it.</title><link>https://willhackett.com/en-us/workforce-engineering/</link><guid isPermaLink="true">https://willhackett.com/en-us/workforce-engineering/</guid><description>Every discipline starts the same way. People do the work for years before anyone thinks to define it. Workforce Engineering is the discipline of deliberately designing, measuring and optimizing how an organization deploys its labor to produce outcomes.
</description><pubDate>Sat, 11 Apr 2026 14:35:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><category>product</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>What is Workforce Engineering, and how do you achieve flow state?
</title><link>https://willhackett.com/en-us/what-is-workforce-engineering/</link><guid isPermaLink="true">https://willhackett.com/en-us/what-is-workforce-engineering/</guid><description>Workforce Engineering is the practice of deliberately designing, measuring and optimizing how an organization deploys its labor, human and AI, to produce outcomes. Here&apos;s the framework, the six practices, and how to get started.
</description><pubDate>Sat, 11 Apr 2026 08:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><category>product</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Boring is better: Pushing 67k telemetry events per second into Postgres</title><link>https://willhackett.com/en-us/just-use-postgres/</link><guid isPermaLink="true">https://willhackett.com/en-us/just-use-postgres/</guid><description>I needed to store massive OpenTelemetry data for AI spend analysis. Instead of reaching for BigQuery or Snowflake, I pushed 67,000 telemetry events per second into Postgres on a MacBook Air, while running analytical queries against it, to prove the boring choice scales.</description><pubDate>Sun, 22 Mar 2026 07:49:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><category>infrastructure</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Amortizing a hallucination</title><link>https://willhackett.com/en-us/amortising-a-hallucination/</link><guid isPermaLink="true">https://willhackett.com/en-us/amortising-a-hallucination/</guid><description>We are currently adjusting our EBITDA to account for a sophisticated parrot that read the internet. Gartner says $2.52 trillion in global AI spend this year, yet only 14% of CFOs report clear ROI. Sequoia&apos;s 6:1 services-to-software ratio means the problem is about to get six times worse.
</description><pubDate>Fri, 13 Mar 2026 01:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>ai</category><category>business</category><category>finance</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>AI agents won&apos;t kill your system of record. Your rate limits will.</title><link>https://willhackett.com/en-us/ai-sor/</link><guid isPermaLink="true">https://willhackett.com/en-us/ai-sor/</guid><description>AI agents won&apos;t hollow out systems of record. But the platforms that fight agent access with rate limits instead of embracing structured protocols like MCP will lose to the ones that don&apos;t.
</description><pubDate>Sun, 08 Feb 2026 21:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>ai</category><category>business</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Digital ID done right: verifiable claims without the surveillance</title><link>https://willhackett.com/en-us/age-verification-digital-id/</link><guid isPermaLink="true">https://willhackett.com/en-us/age-verification-digital-id/</guid><description>We have a digital identity problem. Not because verification technology can&apos;t inherently be private, it can. But, because governments are asking the wrong question. Governments ask &quot;how do we identify people?&quot; when what they should be asking is &quot;how do we verify claims without identifying anyone?&quot;
</description><pubDate>Tue, 03 Feb 2026 21:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Why I moved to Microsoft 365</title><link>https://willhackett.com/en-us/fastmail-to-microsoft/</link><guid isPermaLink="true">https://willhackett.com/en-us/fastmail-to-microsoft/</guid><description>After years with Fastmail, we switched to Microsoft 365. Spam filtering, EU data residency, and ecosystem integration made the difference.
</description><pubDate>Mon, 12 Jan 2026 08:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>business</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Digital sovereignty requires funding, not just adoption</title><link>https://willhackett.com/en-us/eu-open-source-sovereignty/</link><guid isPermaLink="true">https://willhackett.com/en-us/eu-open-source-sovereignty/</guid><description>The EU sees open source as the path to digital sovereignty. They&apos;re right. But adoption without sustainable funding just means Big Tech keeps extracting value from European developers.</description><pubDate>Sat, 10 Jan 2026 09:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>opinion</category><category>business</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>The best engineering teams can turn on a dime</title><link>https://willhackett.com/en-us/engineering-teams-turn-on-a-dime/</link><guid isPermaLink="true">https://willhackett.com/en-us/engineering-teams-turn-on-a-dime/</guid><description>Why fluid planning beats rigid quarters, and how the best CTOs build cultures that adapt without chaos.
</description><pubDate>Mon, 05 Jan 2026 12:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>business</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Agentic coding is changing the engineering workforce, just not how you think</title><link>https://willhackett.com/en-us/agentic-coding-workforce/</link><guid isPermaLink="true">https://willhackett.com/en-us/agentic-coding-workforce/</guid><description>AI coding tools have crossed a critical adoption threshold. But the real shift isn&apos;t individual productivity. It&apos;s how engineering organizations allocate capacity across priorities.
</description><pubDate>Wed, 24 Dec 2025 09:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>ai</category><category>business</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Reclaiming the commons: the case for an accountable internet</title><link>https://willhackett.com/en-us/reclaiming-the-commons/</link><guid isPermaLink="true">https://willhackett.com/en-us/reclaiming-the-commons/</guid><description>On the case for moving away from anonymous &quot;ghost&quot; accounts toward a pragmatic internet. One where we can protect the vulnerable without silencing the dissident.</description><pubDate>Sun, 21 Dec 2025 12:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>opinion</category><category>society</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Moore&apos;s Law for AI is officially dead</title><link>https://willhackett.com/en-us/cost-of-ai-inference/</link><guid isPermaLink="true">https://willhackett.com/en-us/cost-of-ai-inference/</guid><description>Exploring historic AI model prices, and Google&apos;s new Gemini 3 Flash at $0.50/$3.00 per million tokens, a 67% increase over 2.5 Flash.
</description><pubDate>Wed, 17 Dec 2025 22:20:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><category>ai</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Big teams, small team energy</title><link>https://willhackett.com/en-us/big-teams-small-team-energy/</link><guid isPermaLink="true">https://willhackett.com/en-us/big-teams-small-team-energy/</guid><description>It&apos;s IEEE Spectrum time again, so it&apos;s time to talk about why adding management layers early is the real problem behind $2 trillion in annual software failures. From Canada&apos;s Phoenix payroll system to the UK Post Office&apos;s Horizon fiasco.
</description><pubDate>Tue, 25 Nov 2025 12:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>business</category><category>technical</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Search kind of sucks now</title><link>https://willhackett.com/en-us/ai-search-crisis/</link><guid isPermaLink="true">https://willhackett.com/en-us/ai-search-crisis/</guid><description>Google&apos;s AI Overview recently told users that astronauts met cats on the moon. It suggested adding glue to pizza to stop cheese sliding off. For a brief period in early 2025, it claimed we were still in 2024. These aren&apos;t edge cases. They&apos;re symptoms of a much larger problem that&apos;s quietly dismantling the internet&apos;s economic foundation. LLMs are fundamentally breaking the internet&apos;s business model.
</description><pubDate>Sat, 15 Nov 2025 02:31:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><author>hello@willhackett.com (Will Hackett)</author></item><item><title>How do you explain the £20M engineering spend?
</title><link>https://willhackett.com/en-us/making-flowstate/</link><guid isPermaLink="true">https://willhackett.com/en-us/making-flowstate/</guid><description>So you&apos;re spending 20 million on engineering. What are you getting for it? Software engineering organizations struggle to answer basic questions about costs, and PE scrutiny is changing the game. It&apos;s more important now than ever to treat engineering like the investment it is.
</description><pubDate>Wed, 05 Nov 2025 15:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><category>product</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Fixing email spam with AI: When traditional filters aren&apos;t enough</title><link>https://willhackett.com/en-us/fixing-email-spam-with-ai/</link><guid isPermaLink="true">https://willhackett.com/en-us/fixing-email-spam-with-ai/</guid><description>89% of malicious emails bypass SPF, DKIM and DMARC. Your inbox is already compromised. Authentication protocols failed and nobody told you. Here&apos;s how AI catches what traditional filters miss.</description><pubDate>Sun, 02 Nov 2025 19:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><category>ai</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Pragmatic: Lessons from building (and winding down) an AI startup</title><link>https://willhackett.com/en-us/pragmatic-sunset/</link><guid isPermaLink="true">https://willhackett.com/en-us/pragmatic-sunset/</guid><description>Reflections on building Jamie, an AI assistant for email, and the decision to wind down Pragmatic after not finding product-market fit.
</description><pubDate>Fri, 17 Oct 2025 00:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>business</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Analytics is fine, tracking is the problem</title><link>https://willhackett.com/en-us/analytics-fine-tracking-problem/</link><guid isPermaLink="true">https://willhackett.com/en-us/analytics-fine-tracking-problem/</guid><description>Analytics is curiosity about what works. Tracking is surveillance. I run Umami on Vercel with Neon Postgres for zero-cost, banner-free analytics. The irony? My joke LLM needs consent, but tracking 16,000 visitors doesn&apos;t.
</description><pubDate>Sun, 12 Oct 2025 11:30:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>AI everywhere is AI nowhere</title><link>https://willhackett.com/en-us/ai-everywhere/</link><guid isPermaLink="true">https://willhackett.com/en-us/ai-everywhere/</guid><description>The tech industry&apos;s rush to embed AI into every product is creating exhausted users and half-baked features. Real AI success comes from solving specific problems, not sprinkling chatbots everywhere.
</description><pubDate>Tue, 30 Sep 2025 19:50:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>ai</category><category>opinion</category><category>technical</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Escaping the AI cost trap</title><link>https://willhackett.com/en-us/ai-economics/</link><guid isPermaLink="true">https://willhackett.com/en-us/ai-economics/</guid><description>AI is cheap for end-users right now, but only because venture capital is picking up the tab. The reality is that training and inference are still expensive, and someone has to pay. Billions raised by OpenAI, Anthropic, Mistral and others are spent on compute, staff and Nvidia GPUs. Consumers see subsidized pricing, but the underlying costs are enormous.
</description><pubDate>Fri, 26 Sep 2025 22:52:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>ai</category><category>business</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Web 4.0: The Pragmatic Internet</title><link>https://willhackett.com/en-us/web-4-0-pragmatic-internet/</link><guid isPermaLink="true">https://willhackett.com/en-us/web-4-0-pragmatic-internet/</guid><description>My take on the next version of the web given the failures of Web 3.0 and the need for a more organized, user-centric approach.</description><pubDate>Wed, 27 Aug 2025 01:40:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>Technical</category><category>AI</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Berlin Airport reflections: on polymaths, ownership and execution</title><link>https://willhackett.com/en-us/berlin-airport-reflections/</link><guid isPermaLink="true">https://willhackett.com/en-us/berlin-airport-reflections/</guid><description>Reflections on meeting a team of scrappy, multi-faceted engineers and what makes the best startups work
</description><pubDate>Fri, 15 Aug 2025 00:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>business</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>How to Not Decrypt WhatsApp Web (But Still Win)</title><link>https://willhackett.com/en-us/whatsapp-and-tonic/</link><guid isPermaLink="true">https://willhackett.com/en-us/whatsapp-and-tonic/</guid><description>A tale of 6,000 f-bombs and one accidental victory. What started as a simple integration for Jamie turned into a deep dive through WhatsApp&apos;s encryption, the Signal Protocol, and eventually finding a simpler path through DOM scraping.
</description><pubDate>Wed, 06 Aug 2025 23:22:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><category>programming</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Teaching AI to feel: the architecture of emotional intelligence</title><link>https://willhackett.com/en-us/emotional-ai/</link><guid isPermaLink="true">https://willhackett.com/en-us/emotional-ai/</guid><description>What if your AI assistant could actually feel what kind of day you&apos;re having? Not in some creepy sci-fi way, but in a structured, measurable way that makes it respond like a human would, with actual understanding.
</description><pubDate>Tue, 05 Aug 2025 11:44:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>ai</category><category>machine learning</category><category>technical</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Stop reinventing the wheel</title><link>https://willhackett.com/en-us/avoid-reinventing-the-wheel/</link><guid isPermaLink="true">https://willhackett.com/en-us/avoid-reinventing-the-wheel/</guid><description>How switching from a broken WordPress site to Luma transformed It&apos;s Culture Innit&apos;s event management, proving that buying beats building when software isn&apos;t your core value.</description><pubDate>Thu, 10 Jul 2025 23:45:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>business</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Know your Geebos</title><link>https://willhackett.com/en-us/know-your-geebos/</link><guid isPermaLink="true">https://willhackett.com/en-us/know-your-geebos/</guid><description>Essential principles for defining Generic Business Objects (GBOs) in your system architecture to avoid expensive re-engineering and maintain clean data models.</description><pubDate>Fri, 20 Jun 2025 21:38:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><category>programming</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Don&apos;t take the cloud call - focus on your startup</title><link>https://willhackett.com/en-us/the-cloud-call/</link><guid isPermaLink="true">https://willhackett.com/en-us/the-cloud-call/</guid><description>A founder&apos;s warning about why early-stage startups should ignore cloud vendor sales calls and keep infrastructure simple to preserve runway and focus on shipping product.</description><pubDate>Thu, 19 Jun 2025 20:20:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>business</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Your website is pissing people off</title><link>https://willhackett.com/en-us/captcha-pissing-off-old-people/</link><guid isPermaLink="true">https://willhackett.com/en-us/captcha-pissing-off-old-people/</guid><description>How unnecessary CAPTCHAs, cookie prompts, and popup modals are driving customers away from websites, with practical advice on reducing friction.</description><pubDate>Thu, 29 May 2025 20:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>business</category><category>opinion</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Invest in integrations, not the platforms</title><link>https://willhackett.com/en-us/invest-in-integration/</link><guid isPermaLink="true">https://willhackett.com/en-us/invest-in-integration/</guid><description>Why VCs should back focused tools that integrate well with existing platforms instead of chasing the next monolithic unicorn that tries to do everything.</description><pubDate>Thu, 08 May 2025 16:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>business</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>How to actually get things built</title><link>https://willhackett.com/en-us/how-to-actually-get-things-built/</link><guid isPermaLink="true">https://willhackett.com/en-us/how-to-actually-get-things-built/</guid><description>Why product teams need to embrace simplicity over complexity, focusing on clear blueprints and agility rather than endless documentation and rigid processes.</description><pubDate>Fri, 02 May 2025 16:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>business</category><category>technical</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>UK Pensions just aren&apos;t as good as Australian Superannuation</title><link>https://willhackett.com/en-us/uk-vs-aus-pensions/</link><guid isPermaLink="true">https://willhackett.com/en-us/uk-vs-aus-pensions/</guid><description>A detailed comparison of UK pensions versus Australian superannuation, highlighting how Australia&apos;s mandatory 11% employer contributions, lower fees, and better returns outshine the UK&apos;s restrictive system.</description><pubDate>Sat, 29 Mar 2025 09:03:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>opinion</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Getting started today</title><link>https://willhackett.com/en-us/getting-started/</link><guid isPermaLink="true">https://willhackett.com/en-us/getting-started/</guid><description>A reflection on how web development has evolved from the simple FTP upload days to today&apos;s complex ecosystem of frameworks, security concerns, and over-engineering.</description><pubDate>Fri, 28 Feb 2025 09:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>opinion</category><category>technical</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>I think BunnyCDN is the first real European answer to Cloudflare</title><link>https://willhackett.com/en-us/bunny/</link><guid isPermaLink="true">https://willhackett.com/en-us/bunny/</guid><description>A review of BunnyCDN as a European alternative to Cloudflare, highlighting its strengths in Edge Rules and constraint-driven development approach.</description><pubDate>Wed, 26 Feb 2025 20:26:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><category>opinion</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>The pitfalls of iCloud Mail - systematic failures and user hostility</title><link>https://willhackett.com/en-us/apple-email-is-not-good/</link><guid isPermaLink="true">https://willhackett.com/en-us/apple-email-is-not-good/</guid><description>A detailed investigation into iCloud Mail&apos;s filtering failures, broken rules system, and unresponsive support that led to missed contracts and a migration to Fastmail.</description><pubDate>Tue, 25 Feb 2025 23:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>The cultural cost of a bad hire</title><link>https://willhackett.com/en-us/one-bad-hire/</link><guid isPermaLink="true">https://willhackett.com/en-us/one-bad-hire/</guid><description>The worst kind of bad hire isn&apos;t the one who does nothing, it&apos;s the one who creates chaos to justify their existence. A single bad hire can become cultural rot, fracturing teams, slowing delivery, and driving away top talent through bureaucratic chaos and toxic behavior. They make your team turn off.</description><pubDate>Fri, 21 Feb 2025 01:12:43 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>business</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Tinfoil hat moment this evening, could Cloudflare be an NSA front?</title><link>https://willhackett.com/en-us/tinfoil-hat/</link><guid isPermaLink="true">https://willhackett.com/en-us/tinfoil-hat/</guid><description>A tongue-in-cheek conspiracy theory pondering whether Cloudflare&apos;s ubiquitous presence as a CDN, DNS provider, and certificate authority could be the perfect cover for surveillance.</description><pubDate>Mon, 10 Feb 2025 22:41:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>opinion</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>I think B2C is the best way to a successful B2B product</title><link>https://willhackett.com/en-us/b2c-product-thoughts/</link><guid isPermaLink="true">https://willhackett.com/en-us/b2c-product-thoughts/</guid><description>Why the most successful B2B SaaS products like Canva, Shopify, and Notion started as beloved consumer tools before conquering enterprise markets.</description><pubDate>Sat, 08 Feb 2025 20:25:26 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>business</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Friday night code sessions - a lazy CleanShot X alternative</title><link>https://willhackett.com/en-us/cloudwrap/</link><guid isPermaLink="true">https://willhackett.com/en-us/cloudwrap/</guid><description>Building a DIY screenshot sharing solution using Cloudflare R2, Workers, and Siri Shortcuts as an alternative to CleanShot X&apos;s cloud features.</description><pubDate>Fri, 07 Feb 2025 22:20:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><category>programming</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>It&apos;s 2025, and JIRA is still clunky</title><link>https://willhackett.com/en-us/jira-in-2025/</link><guid isPermaLink="true">https://willhackett.com/en-us/jira-in-2025/</guid><description>A frustrated rant about JIRA&apos;s clunky interface and poor user experience after switching back from Linear&apos;s elegant project management.</description><pubDate>Mon, 03 Feb 2025 14:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><category>business</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Operational transformation &amp; conflict-free resolution for real-time collaboration applications</title><link>https://willhackett.com/en-us/operational-transformation/</link><guid isPermaLink="true">https://willhackett.com/en-us/operational-transformation/</guid><description>Recently talking with a company that needs to provide real-time collaboration within their application, I started to think about the problem. Operational transformation (OT) and conflict-free replicates data types (CRDTs) are solutions to this problem.
</description><pubDate>Wed, 01 May 2024 05:45:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Techniques to make rebasing less of a headache</title><link>https://willhackett.com/en-us/rebase-techniques/</link><guid isPermaLink="true">https://willhackett.com/en-us/rebase-techniques/</guid><description>I often see developers merging main into their feature branches and opening pull requests (PRs) that include commits from other branches. This practice leads to &quot;git spaghetti,&quot; which has several consequences. But it&apos;s easy to learn to be better.</description><pubDate>Sun, 24 Mar 2024 05:52:40 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>git</category><category>programming</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>I built a GPT-powered email assistant - Jamie</title><link>https://willhackett.com/en-us/jamie-ai/</link><guid isPermaLink="true">https://willhackett.com/en-us/jamie-ai/</guid><description>Jamie is a project I embarked on to build a human-like virtual assistant that I could CC on emails to take care of my admin. It was originally going to be Open Source, but I&apos;ve decided to make it into a product.</description><pubDate>Thu, 15 Feb 2024 14:00:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>machine learning</category><category>edge compute</category><category>gpt</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Sharing encrypted data over short-form mediums</title><link>https://willhackett.com/en-us/share-encrypted-short-form-data/</link><guid isPermaLink="true">https://willhackett.com/en-us/share-encrypted-short-form-data/</guid><description>Sharing large amounts of data over short-form mediums like QR codes or NFC tags can be challenging. The data needs to be encrypted to ensure confidentiality, but packing useful data into a small space is difficult.</description><pubDate>Mon, 24 Jul 2023 12:02:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>programming</category><category>security</category><category>encryption</category><category>cryptography</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Distributed Trust with Service-to-Service Authentication</title><link>https://willhackett.com/en-us/s2s-trust/</link><guid isPermaLink="true">https://willhackett.com/en-us/s2s-trust/</guid><description>In Service-oriented Architectures (SOA) the &quot;services&quot; are the fundamental building blocks of your application. Authenticating requests between services is a challenge.</description><pubDate>Fri, 14 Jul 2023 22:30:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>programming</category><category>security</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>OCR on Edge Compute</title><link>https://willhackett.com/en-us/edge-ocr/</link><guid isPermaLink="true">https://willhackett.com/en-us/edge-ocr/</guid><description>Running models closer to your users can be a huge win for performance and cost. Here&apos;s an experiment in running OCR on the edge with Cloudflare Constellation.</description><pubDate>Sun, 02 Jul 2023 22:30:00 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>machine learning</category><category>programming</category><category>edge compute</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Quit nesting React providers</title><link>https://willhackett.com/en-us/react-compose/</link><guid isPermaLink="true">https://willhackett.com/en-us/react-compose/</guid><description>Straightforward solution for composing a bunch of React Provider components to avoid excessive nesting.</description><pubDate>Tue, 10 May 2022 01:12:43 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>snippet</category><category>programming</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Save time importing Cloudflare DNS records</title><link>https://willhackett.com/en-us/cloudflare-bind/</link><guid isPermaLink="true">https://willhackett.com/en-us/cloudflare-bind/</guid><description>Cloudflare&apos;s UI for managing DNS records is nice, but this method can help you import larger batches of changes faster.</description><pubDate>Mon, 11 Apr 2022 01:12:43 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>snippet</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Rebase to freshen up your branch</title><link>https://willhackett.com/en-us/git-rebase/</link><guid isPermaLink="true">https://willhackett.com/en-us/git-rebase/</guid><description>I&apos;m a member of a few software teams and, very frequently, we commit code to our main branch. Common themes within my teams are not to have a develop branch or even release cycles. I practice rebasing instead of merging to keep my branch fresh.</description><pubDate>Sun, 18 Jul 2021 01:12:43 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>technical</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Apple could do a lot with distributed compute</title><link>https://willhackett.com/en-us/distributed-compute/</link><guid isPermaLink="true">https://willhackett.com/en-us/distributed-compute/</guid><description>Being an avid user of Apple&apos;s products, one of the standout features that I enjoy is the on-device processing of my personal data. Be it searching for an image, or Siri suggesting a calendar event based on a flight in my email, it&apos;s extremely useful.</description><pubDate>Fri, 09 Jul 2021 01:12:43 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>opinion</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Grocery Store coding challenge</title><link>https://willhackett.com/en-us/grocery-store/</link><guid isPermaLink="true">https://willhackett.com/en-us/grocery-store/</guid><description>This coding challenge is part of a series I will be posting for TypeScript &amp; Golang, although it can be completed in other languages. As these are used for internal recruitment purposes, I recommend that you keep repositories with this challenge.</description><pubDate>Wed, 07 Jul 2021 01:12:43 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>programming</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Token invalidation in distributed systems</title><link>https://willhackett.com/en-us/distributed-token-invalidation/</link><guid isPermaLink="true">https://willhackett.com/en-us/distributed-token-invalidation/</guid><description>Building fast systems can be tricky and distributed computing is often the approach taken to improve performance.</description><pubDate>Sat, 08 Feb 2020 01:12:43 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>security</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Unit testing in Keystone JS</title><link>https://willhackett.com/en-us/unit-testing-in-keystone-js/</link><guid isPermaLink="true">https://willhackett.com/en-us/unit-testing-in-keystone-js/</guid><description>How to implement Mocha &amp; Chai unit testing in Keystone JS.</description><pubDate>Thu, 30 Jun 2016 11:12:43 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>javascript</category><category>programming</category><author>hello@willhackett.com (Will Hackett)</author></item><item><title>Let&apos;s Encrypt - time to get serious about TLS</title><link>https://willhackett.com/en-us/lets-encrypt/</link><guid isPermaLink="true">https://willhackett.com/en-us/lets-encrypt/</guid><description>Let&apos;s Encrypt is a new Certificate Authority that&apos;s serious about democratizing SSL on the internet. The team at Let&apos;s Encrypt is making TLS more accessible by validating certificates without the need to pay the likes of Comodo or DigiCert.</description><pubDate>Sun, 10 Apr 2016 01:12:43 GMT</pubDate><dc:creator>Will Hackett</dc:creator><category>security</category><author>hello@willhackett.com (Will Hackett)</author></item></channel></rss>