[{"data":1,"prerenderedAt":32},["ShallowReactive",2],{"$fjMIH3STkm2t8JwjjPIJBc5kShxEfdlfmhx5Un_AEtE4":3,"$fuq8P6VQNTQL6z3cCxpSmowHsPTNGZDi3yQOzq4k3_g4":10,"$fwEyCYBANdkWYIXHzjjlkFH3VzrsbxVP5jVGisJQDFLk":18,"$fCkWELwI2OGQEgV8ZTT-D4N5xOShj5KPSyC9pYJ1wev4":25,"$fiTbj2uX6djAoN9bor_yaIrGEGrVroA-KNcpibaHwjfs":29},[4,7],{"slug":5,"url":6},"knowing_claude_part_1","\u002Fapi\u002Fcontent\u002Fknowing_claude_part_1",{"slug":8,"url":9},"about","\u002Fapi\u002Fcontent\u002Fabout",{"slug":5,"frontmatter":11,"body":16,"preview":17},{"title":12,"category":13,"date":14,"image":15},"Getting to know Anthropic's Claude","Technology; AI","31 May 2026",".\u002Fassets\u002FClaude_Sonnet_4.6.png","\u003Ch2>Intro\u003C\u002Fh2>\n\u003Cp>Being an 'old skool' developer, one so that just has to get hands-on whenever possible, I find myself immersed in an environment that only leads in one direction - towards an Artificial Intelligence (AI) driven world.  There is no opting-out it seems given the significant investment by every company out there trying to outgun each other by leveraging this technology.  It's funny, it does of come with many compromises in its early stages of usage and I have seem a good few examples of these.\u003C\u002Fp>\n\u003Cp>To start off then in my quest to gain AI literacy and develop my AI skills, I created myself two projects.  One project is where I am aiming to develop a specific example software product, one that is fully documented and is quite strict in its implementation and the the other project - this blog and website - where I can simply sit back, relax and vibe code it.  The site does not have to be perfect, so I will as someone who treats quality with respect, I will still try to perfect it!  In adopting this approach to interact with AI I have found is quite an effective way of understanding where it makes sense to use it and where fundamental and core foundational skills can still be significantly better.\u003C\u002Fp>\n\u003Cp>On to talk about this site then ...\u003C\u002Fp>\n\u003Ch2>How this website was born\u003C\u002Fh2>\n",true,{"slug":8,"frontmatter":19,"body":24,"preview":17},{"title":20,"category":21,"date":22,"image":23},"5, 4, 3, 2, 1 … So here we go!","General","16 May 2026",".\u002Fassets\u002Fbanner.png","\u003Cp>Welcome to 37hoursaday.com! \u003C\u002Fp>\n\u003Cp>I have always wanted to set up and write a blog so that I can share my thoughts, mainly on technology, and maybe other subject matters as they become apparent or if they are noteworthy enough. \u003C\u002Fp>\n\u003Cp>So here it is... \u003C\u002Fp>\n\u003Cp>I do not know at this time whether what I write will be insightful, funny or otherwise, however it will be nice if it gets you thinking… \u003C\u002Fp>\n\u003Cp>The site itself (not the content) is completely built of a ‘vibe coded’ implementation that shows with sensible inputs you can produce reasonable and acceptable results albeit with some negotiation.  I have been using Anthropic’s Claude Sonnet 4.6 to do this even though there are seemingly more capable models available at the time of writing.  I will be sure to blog about this in more detail as time goes on to describe how I [\u003Cdel>embraced\u003C\u002Fdel> == started with] and [\u003Cdel>adopted\u003C\u002Fdel> == carried on using] this AI 'thing'.\u003C\u002Fp>\n",{"slug":5,"frontmatter":26,"body":27,"preview":28},{"title":12,"category":13,"date":14,"image":15},"\u003Ch2>Intro\u003C\u002Fh2>\n\u003Cp>Being an 'old skool' developer, one so that just has to get hands-on whenever possible, I find myself immersed in an environment that only leads in one direction - towards an Artificial Intelligence (AI) driven world.  There is no opting-out it seems given the significant investment by every company out there trying to outgun each other by leveraging this technology.  It's funny, it does of come with many compromises in its early stages of usage and I have seem a good few examples of these.\u003C\u002Fp>\n\u003Cp>To start off then in my quest to gain AI literacy and develop my AI skills, I created myself two projects.  One project is where I am aiming to develop a specific example software product, one that is fully documented and is quite strict in its implementation and the the other project - this blog and website - where I can simply sit back, relax and vibe code it.  The site does not have to be perfect, so I will as someone who treats quality with respect, I will still try to perfect it!  In adopting this approach to interact with AI I have found is quite an effective way of understanding where it makes sense to use it and where fundamental and core foundational skills can still be significantly better.\u003C\u002Fp>\n\u003Cp>On to talk about this site then ...\u003C\u002Fp>\n\u003Ch2>How this website was born\u003C\u002Fh2>\n\u003Cp>I see this website as being dynamic in that I will continue to vibe code it, changing it radically or adding new features as it evolves or if the mood takes me.  It's interesting, when I first started out thinking about how to create the site I was sat with that familiar blank sheet pondering what would be possible.  As a full stack polyglot developer I could make any choices I wanted, unshackled from compliance, committees and working groups.  Guess what? Even with an infinite number of options I still defaulted to my favourite frontend stack, one that I knew certainly had advanced more than when I last used it so there was a good chance to learn something.  As would the case, I set off, used the CLI and started to develop some functionality.\u003C\u002Fp>\n\u003Cp>Little did I know at this point that I would start to think how to utilise AI, but then I did ...\u003C\u002Fp>\n\u003Ch2>Getting to know Anthropic's Claude\u003C\u002Fh2>\n\u003Cp>Having researched, and read an endless number of articles on which AI could potentially work for me, I opted to just pick one ...  In this case I opted for Anthropic Claude, to set up a Pro subscription and then use it to the maximum extent possible.  In the early stages I wanted to use the AI pretty much like any other tool, to step through and understand what it is doing point to point.  As I started to become more familiar with Claude I started to think of it as pair programmer, one that could clearly know more than me, yet could still accept me as a fellow developer.  This really did take some getting used to.\u003C\u002Fp>\n\u003Cp>Given time is now progressing I found it more and more acceptable for Claude to achieve things for me.  I forced myself not to make any code level changes and through prompting alone had Claude do the heavy lifting.  Notably I will add that given the early quality level of this site, I am quite happy for Claude to have a number of goes to get things right or in to an acceptable state.  \u003C\u002Fp>\n\u003Ch3>Taking that leap to cross 'the divide' into the AI dimension\u003C\u002Fh3>\n\u003Cp>Having used Claude now for around two weeks (at the time of writing), I have made the quick transition to mean now that there will likely be some level of AI interactions in my day to day coding activities.  I think this is mostly aided through the making of this site given I have not got any guardrails to apply in its development.  When I finally decided to allow for Claude to 'Edit automatically' I felt a real sense of achievement whereby even in the short time I had used it [and the risk I was taking] I had started to trust it.  I can now set a task and quite confidently walk away and come back when it has finished.  Once any developer achieves this no matter how sceptical you are it does provide a real advantage. Lets see how far this gets me ...\u003C\u002Fp>\n\u003Cp align=\"center\">\n  \u003Cimg src=\".\u002Fassets\u002Fvscode_claude_edit_auto.png\" alt=\"Description\" width=\"200\" height=\"50\">\n\u003C\u002Fp>\n\n\u003Ch2>Key Takeaways so far\u003C\u002Fh2>\n\u003Cul>\n\u003Cli>Core foundational software engineering principles still matter, never ever forget this.\u003C\u002Fli>\n\u003Cli>If more tasks, even complex ones are automated, then it frees up time to be more creative.\u003C\u002Fli>\n\u003Cli>My way of thinking how software can be developed has already and permanently changed.  I for one am very relieved how this transition has been quick, easy, painless and mostly seamless.\u003C\u002Fli>\n\u003Cli>Thinking of AI as a tool is a very sensible stance to take in the short term.  It can solve problems, it can create problems, but it can still be a very handy tool to learn from.\u003C\u002Fli>\n\u003Cli>As with any technology it is best to use AI on a very regular basis to keep all your skills fresh and relevant.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2>In Summary\u003C\u002Fh2>\n\u003Cp>Throughout the last few months, my thoughts around AI, its controversies, its supposed benefits and everything else that comes with the use of it have definitely changed significantly.  I have always been very adaptive to change and continue to embrace it.  The speed at which this change is happening though is what we need to adapt to not what the technology is currently doing or what it could do in the future.  I will keep you updated in future posts on how my opinion and perspective changes on my journey to the AI enabled future.\u003C\u002Fp>\n",false,{"slug":8,"frontmatter":30,"body":31,"preview":28},{"title":20,"category":21,"date":22,"image":23},"\u003Cp>Welcome to 37hoursaday.com! \u003C\u002Fp>\n\u003Cp>I have always wanted to set up and write a blog so that I can share my thoughts, mainly on technology, and maybe other subject matters as they become apparent or if they are noteworthy enough. \u003C\u002Fp>\n\u003Cp>So here it is... \u003C\u002Fp>\n\u003Cp>I do not know at this time whether what I write will be insightful, funny or otherwise, however it will be nice if it gets you thinking… \u003C\u002Fp>\n\u003Cp>The site itself (not the content) is completely built of a ‘vibe coded’ implementation that shows with sensible inputs you can produce reasonable and acceptable results albeit with some negotiation.  I have been using Anthropic’s Claude Sonnet 4.6 to do this even though there are seemingly more capable models available at the time of writing.  I will be sure to blog about this in more detail as time goes on to describe how I [\u003Cdel>embraced\u003C\u002Fdel> == started with] and [\u003Cdel>adopted\u003C\u002Fdel> == carried on using] this AI 'thing'.\u003C\u002Fp>\n\u003Cp>The logo for the site if you look closely is, yep is a ‘37 hour’ clock.  Not because we are on a planet that revolves more slowly than Earth, more so as the world continues on its current path, we are all seemingly expected to do far more with the same or much less – whether this be related to people, skills or computational capabilities or resources.  Even with AI in the picture this is now very unlikely to change back.  The other hidden elements in the logo I will keep for another time, perhaps changing with my mood or for different reasons altogether.\u003C\u002Fp>\n\u003Cp>Welcome to 37hoursaday.com! \u003C\u002Fp>\n",1780248372923]