Lloyd Jones @lloydjones.io · 19/05/2026systima.ai/blog/continu...systima.aiContinuous Conformity: Engineering Evidence for Orchestrated AI SystemsApplied research note proposing continuous conformity as the engineering response to the orchestration-layer governance gap identified in The Hidden Layer (EU Law Live, 2026). 000
Lloyd Jones @lloydjones.io · 01/04/2026What the Claude Code Leak Means for Regulated Industries systima.ai/blog/claude-...systima.aiWhat The Claude Code Leak Means for Engineering Teams in Regulated IndustriesThe leaked codebase reveals engineering practices that should inform how regulated teams assess their AI toolchain dependencies. 110
Lloyd Jones @lloydjones.io · 23/03/2026The fractional Head of AI market has two models: brokers who deploy generalists, and practitioners who've built and shipped AI systems in production. Six questions to tell which you're getting: systima.ai/blog/what-to... #AILeadership #HeadOfAI #AIStrategy #AgenticAI #StartupCTOsystima.aiWhat to Look for in a Fractional Head of AI (and What to Avoid)The fractional AI leadership market has two models: brokers who deploy generalists and practitioners who make architectural decisions. Six questions reveal which you are getting. 000
Lloyd Jones @lloydjones.io · 23/03/2026Saving Opus 4.6 tokens without losing quality #claude #ai www.linkedin.com/posts/lloyda...linkedin.com#aiengineering #developerproductivity #agenticcoding #guidedexecutionpattern | Lloyd J.Something that's been working well for us and might save others some money (summarised with AI): We use Open Code (switched from Claude Code because it lets you mix models). The workflow we've landed... 010
Lloyd Jones @lloydjones.io · 20/03/2026A leaked Google Sheet just exposed how Delve, a Y Combinator-backed GRC platform with $32M in Series A funding, generated 494 near-identical SOC 2 audit reports. They used the same boilerplate, and the same "independent" conclusions systima.ai/blog/delve-c... #delve #aisystima.aiDelve Faked 494 Compliance Reports. The EU AI Act Was Designed to Prevent Exactly This.A leaked spreadsheet revealed that GRC platform Delve generated 494 near-identical SOC 2 reports with pre-written auditor conclusions. The EU AI Act's conformity assessment framework was designed to p... 000
Lloyd Jones @lloydjones.io · 14/03/2026A CLI tool that scans your codebase for EU AI Act compliance risks and optionally blocks deployment if any are detected. Saves them as a PR comment and/or PDF. systima.ai/blog/systima... #euaiact #github #agenticsystima.aiSystima Comply: EU AI Act Compliance Scanning for CI/CD PipelinesSystima Comply: open-source EU AI Act compliance scanning for CI/CD. Detects 37+ AI frameworks, validates risk classifications, checks Articles 5-50, and posts findings in pull requests. 030
Lloyd Jones @lloydjones.io · 06/03/2026Systima tested XML vs MD vs JSON prompts across 4 frontier models, 10 tasks, 600 calls. Delimiter format almost never matters. But they found a reproducible 20% prompt injection vulnerability in one model+format combo. That DOES matter. systima.ai/blog/delimit...systima.aiThe Delimiter Hypothesis: Does Prompt Format Actually Matter?We tested XML, Markdown, and JSON delimiters across four frontier LLMs with 600 model calls. For three of four models, format does not matter. For MiniMax M2.5, Markdown has a measurable prompt inject... 000
Lloyd Jones @lloydjones.io · 06/03/2026It seems that they can parse all quite equally well: systima.ai/blog/delimit...systima.aiThe Delimiter Hypothesis: Does Prompt Format Actually Matter?We tested XML, Markdown, and JSON delimiters across four frontier LLMs with 600 model calls. For three of four models, format does not matter. For MiniMax M2.5, Markdown has a measurable prompt inject... 000
Lloyd Jones @lloydjones.io · 03/03/2026systima.ai/blog/eu-ai-a...systima.aiEngineering Leader's Guide to the EU AI ActTranslating Articles 9-15 into architecture, pipelines, and production controls. A comprehensive engineering guide to EU AI Act compliance. 010
Lloyd Jones @lloydjones.io · 25/02/2026Why Many UK Firms Will Need to Comply with the EU AI Act systima.ai/blog/why-uk-...systima.aiWhy Many UK Firms Will Need to Comply with the EU AI ActThe EU AI Act probably applies to your UK business. Here's why. 000
Lloyd Jones @lloydjones.io · 24/11/2025AI agents can break GDPR and the EU AI Act by just doing their job. The law assumes static systems, but agents can be dynamic. I wrote more about this here: ⚖️https://www.europeanlawblog.eu/pub/dq249o3c/ 010
Lloyd Jones @lloydjones.io · 19/11/2025www.youtube.com/watch?v=RKdx...youtube.comCan We Trust Silicon Valley With Superintelligence? — With Nick CleggYouTube video by Alex Kantrowitz 000
Reposted by Lloyd JonesDavid Buchanan @retr0.id · 18/11/2025happy cloudflare outage day to all who celebrate 231504380
Lloyd Jones @lloydjones.io · 08/11/2025pagewatch.ai/blog/post/ll...pagewatch.aiCan you save on LLM tokens using images instead of text?What happens if you convert your prompts to an image before running? 000
Lloyd Jones @lloydjones.io · 07/11/2025Good ideas! Not responding so often is an immediate 'win', and actually I think the care/attention of a longer, more thorough email is often better than a more lazily bashed-out fast response.. 110
Lloyd Jones @lloydjones.io · 07/11/2025Interesting post, but in practice how is one supposed to eliminate the low-velocity activities (social n; personal prod; ref search; calendars — per the example screenshot)? These aren’t (all) pointless indulgences and without them, would the wheels not begin to fall of of work/life? 110
Lloyd Jones @lloydjones.io · 06/11/2025www.youtube.com/watch?v=9GQl...youtube.comTory and Lib Dem chancellors answer questions on Labour's upcoming budgetYouTube video by Sky News 000
Lloyd Jones @lloydjones.io · 05/11/2025www.augmentcode.com/blog/develop...augmentcode.comDevelopers are choosing older AI models — and the data explain whyThe most powerful AI software development platform with the industry-leading context engine. 110
Lloyd Jones @lloydjones.io · 03/11/2025chrisloy.dev/post/2025/08...chrisloy.devContext engineering | Chris LoyAs our use of LLMs has changed from conversational chatbots and into integral decision-making components of complex systems, our inference approach must also evolve. 000
Lloyd Jones @lloydjones.io · 31/10/2025If you can't beat them, join them. Introducing: 🤩 "Visibilii" 🤩 visibilii.vercel.app 000
Lloyd Jones @lloydjones.io · 31/10/2025Onward! Still building the real thing: The world’s first autonomous AI agent for SEO and AEO. ...and still using the original recipe: 11 secret herbs, zero extra vowels. 110
Lloyd Jones @lloydjones.io · 31/10/2025It's very tempting to send them our brand guidelines so they can really get it right. 😉 It's good to see agentic SEO starting to ripple outward – This just isn't in the way I expected! When others start borrowing your language it's usually a sign that you’re setting the pace. 110
Lloyd Jones @lloydjones.io · 31/10/2025I guess our branding is contagious! Someone became 'heavily inspired' by it, added one extra letter to our brand name, and threw it into Loveable.dev for a quick landing page rip-off. Same mission, same tone. 211
Lloyd Jones @lloydjones.io · 31/10/2025karashiiro.leaflet.pub/3m4gf7geefs2lkarashiiro.leaflet.pubA Look at ANTML: The Anthropic Markup Language - Kara's NonsenseA brief trip into the depths of Anthropic's API 000
Lloyd Jones @lloydjones.io · 27/10/2025blog.joemag.dev/2025/10/the-...blog.joemag.devThe New Calculus of AI-based CodingA collection of posts by Joe Magerramov 130
Lloyd Jones @lloydjones.io · 22/10/2025Interesting, I see — Whereas the others are more short-termist and beholden to shareholders? 010
Lloyd Jones @lloydjones.io · 22/10/2025I am curious: Do the others have business models that genuinely couldn’t support higher wages? Or it’s more… a choice? 100
Lloyd Jones @lloydjones.io · 08/10/2025By the 70s/80s, virtually every new or modernised primary school was built with a multipurpose hall that doubled as dining room, assembly hall, and PE space — and the apparatus became a standard feature. 000
Lloyd Jones @lloydjones.io · 08/10/2025• Firms like Nissen, Huck, and Universal Services supplied standard modular wall-mounted climbing frames, foldable for hall space, with attachable ladders, ropes, and beams. • Local Education Authorities (LEAs) often bulk-purchased these to equip new or refurbished schools. 100
Lloyd Jones @lloydjones.io · 08/10/2025Mine (in the 90’s) had this. Summarised from ChatGPT: • The govt’s School Building Programme (and later DES) building standards included specifications for gymnasia and approved apparatus. 110
Reposted by Lloyd JonesReal Food Budget @real-food-budget.bsky.social · 23/09/2025Ever notice how small grocery spends – £5 here, £3 there – somehow add up to a lot by the end of the month? 😳 Tracking your food spend can be really eye-opening. Curious how much you’re really spending? 💬 Reply with your most surprising grocery bill! #GrocerySpend #BudgetTips 001
Reposted by Lloyd JonesMo Ali | مح علي @moali.bsky.social · 19/09/2025If you're on LinkedIn, you'll probably want to switch this toggle to 'off'. 2217393
Lloyd Jones @lloydjones.io · 15/09/2025www.youtube.com/watch?v=-18K...youtube.comBeethoven: Duo No.1 in C major for clarinet and bassoonYouTube video by Academy of Ancient Music 010
Lloyd Jones @lloydjones.io · 23/08/2025harpers.org/archive/2025...harpers.orgNone the Wiser, by Meghan O’Gieblyn 120
Reposted by Lloyd JonesTom Dörr @tom-doerr.bsky.social · 23/08/2025secure .env file management with encryption and multi-environment support 111
Lloyd Jones @lloydjones.io · 16/08/20254/ Early results are proving that businesses can optimise not just for Google, but for how AI systems actually read and answer. Huge thanks to the early adopters who’ve trusted us with their sites; we’re only just getting started. 100
Lloyd Jones @lloydjones.io · 16/08/20253/ The screenshots below show Visibil in action: ✅ Identifying opportunities ✅ Generating concrete, page-level recommendations ✅ Helping customers climb in both traditional search and AI-driven answer engines It’s great to see something we’ve believed in start creating real impact. 100
Lloyd Jones @lloydjones.io · 16/08/20252/ Now, it’s not just theory — it’s delivering results for real customers. 100
Lloyd Jones @lloydjones.io · 16/08/2025🧵 🚀 Exciting milestone for @visibil.bsky.social (visibil.ai) We’ve been hard at work building out our Dashboard for users to see the changes that Visibil makes, enabling them to take control of their SEO and Answer Engine Optimization (AEO). 121
Lloyd Jones @lloydjones.io · 08/08/2025In the course of creating @visibil.bsky.social, I had the need to grab website documentation and the content of other websites (for testing purposes) and PDFs, as markdown. This is a pain to keep on top of, so I threw this together "With AI"™ – I hope it's useful for others markdown-vault.vercel.appmarkdown-vault.vercel.appPrivate Markdown Library for AI Agents by AI Slop Labs (Lloyd Jones)Convert URLs and PDFs to private markdown notes. Copy or share securely with a link. Local storage FTW. 001
Lloyd Jones @lloydjones.io · 04/07/2025— Further reading, if you’re in doubt: - vercel.com/blog/how-were-adapting-s… - seo.ai/blog/does-chatgpt-and-ai-cra… 010
Lloyd Jones @lloydjones.io · 04/07/2025Visibil treats SEO and LLMO as infrastructure, not frontend theatre. 110
Lloyd Jones @lloydjones.io · 04/07/2025It looks good to users, but to AI systems, it may as well not exist. 100
Lloyd Jones @lloydjones.io · 04/07/2025They are not browsers. They do not wait for DOM mutations or hydrate your frontend. If the content is not in the HTML, it likely does not make it into the model’s memory. Client-side injection is a risky illusion. 100
Lloyd Jones @lloydjones.io · 04/07/2025If your content only appears AFTER JavaScript runs, it may never be seen, indexed, embedded, or cited. The reason behind this is that these AI systems rely on fast, structured data. 100
Lloyd Jones @lloydjones.io · 04/07/2025GPTBot, ClaudeBot, and Perplexity’s crawler typically fetch raw HTML and extract context from it. 100