My blog was costing me money to serve 17 visitors a month. That felt stupid. So I moved to Hugo and GitHub Pages in about an hour.
That mind-numbing form I faced today turned into an unexpected experiment in stress-free automation. What should have been another “just punch it out” chore revealed how accessible, fast, and surprisingly satisfying automating everyday tasks can be. Rather than falling back on manual grind, a simple setup and a few smart prompts unlocked a new workflow. With instant feedback guiding each tweak, I shaved hours from my process and sidestepped the pitfalls that used to make automation feel like a full-blown weekend project. The ease of connecting all the right context, transforming it on the fly, and watching a polished result appear made the experience feel more like play than work. ...
Discover seamless code exploration—and what to expect—when using Jupyter workflows inside Cursor When I first stumbled upon Jupyter Notebook support in Cursor, it was hard to get genuinely excited—my early attempts were honestly pretty clunky. Editing .ipynb files felt unreliable, and the workflow didn’t quite “click” the way a true notebook should. I kept bouncing between clumsy interfaces, and it didn’t feel worth the hassle. But recently, something changed. I’m not sure exactly when the update landed, but suddenly, working in notebooks inside Cursor is not just possible, but actually a pleasure. ...
You know those weeks when you glance at the AI headlines and think, “Okay, that’s neat, but it’ll be years before we feel it on the ground?” Then something drops that makes you double-take. That’s exactly how the Samsung Tiny Recursive Model (TRM) news hit me. And just as I was digesting that, the Thinking Machines Lab’s work on deterministic LLM inference showed up. Both are challenging what we thought AI needed to get better and more reliable. ...
When building AI agents, one of the biggest challenges is tool selection—especially as the number of available tools increases. Evidence from OpenAI’s developer community and technical documentation highlights that agents’ tool selection accuracy tends to drop noticeably with larger toolsets, sometimes causing irrelevant or random tools to be invoked even with carefully designed prompts. OpenAI’s internal discussions note a clear decline in single-tool call accuracy and an uptick in failure modes as the system prompt and available tool list grows, with developer complaints of incorrect or unexpected tool choices despite best practices. ...
Introduction I recently setup and have been playing around with the “Deploy LLM Chatbots with RAG and Databricks Vector Search” demo from Databricks. While not a complete step-by-step guide, this article offers insights into what to expect, the necessary steps, how to fix issues that may arise, and how to maximize your trial experience quickly. Getting access to Databricks You can get a free 2-week trial from Databricks. I used AWS, and if you want to follow along, you will need to do the same. Here are the main things to know for setting up the trial. ...
The Rise of Private Generative AI and the Need for Enterprise Tools As the potential of Generative AI becomes increasingly apparent, more and more organisations are looking to harness its power for their specific use cases. However, deploying AI models in a private, secure, and compliant manner presents significant challenges. While public AI models like OpenAI and Anthropic have demonstrated remarkable capabilities, many organisations require the ability to leverage private and often sensitive data sources and deploy them in a controlled environment. This is especially critical for industries with strict data privacy and compliance requirements, such as healthcare, FSI and Government. Private Generative AI allows companies to leverage the power of AI while ensuring that sensitive data remains secure and confidential. ...
I recently conducted a poll on LinkedIn, asking how often people use AI in their daily lives. The results were interesting, but what struck me the most was the response rate. Out of over 700+ impressions, only 17 people responded. This got me thinking — is it that people don’t care for polls, or is it me, or perhaps most people just don’t care about AI? I believe we should all be looking at everything we do through the lens of “how can I leverage AI here?” This brings me to a problem I recently encountered and how AI helped me solve it. ...
Recently, three owls decided to make their home in our garden. It’s been a hoot (sorry, had to) getting to know these fascinating creatures. My fascination with them sparked the idea for my first ever machine learning application. I had spent the last month or so working my way through the “Practical Deep Learning for Coders” course by Fastai. The field seemed daunting, with its complex mathematical underpinnings and the perception that only tech giants like Google and Facebook could harness its power. However, as I progressed through the course, I discovered that many of my preconceptions were unfounded. ...
Hey there, tech explorers! Welcome to the wild world of AI and Machine Learning — where algorithms rule, data talks, and the future is now! 🚀💻 Join me on this exciting journey as we unravel the mysteries of AI and ML. From self-driving cars to personalised recommendations, we’ll explore the cool stuff these technologies can do. Whether you’re a tech geek or just curious, this blog is your backstage pass to the world of AI and Machine Learning. So buckle up, let’s dive in, and let the digital magic begin! ...