Apple has introduced a new architecture aimed at addressing a long-standing challenge in AI systems that execute autonomous actions. Solo Swift Crafter breaks down how the integration of a "reviewer" agent shifts the focus from error recovery to prevention, offering a proactive safeguard against … [Read more...] about How Apple Quietly Solved the Biggest Risk in AI Agent Workflows
Julian Horsey
Author Archives:
Executive Editor & Co-Founder of Geeky Gadgets
Julian Horsey is the co-founder of Geeky Gadgets and a veteran technology journalist with over 18 years of experience. A published author and multimedia designer, he specializes in emerging frontiers including Artificial Intelligence, autonomous agents, and next-gen hardware like mini PCs and wearables.
Driven by curiosity and a passion for design, Julian provides in-depth analysis on how AI workflows—from Claude to Google Gemini—are transforming productivity and creativity. His work has been cited by global tech leaders and major publications, reflecting nearly two decades of expertise in 3D printing, AR/VR, and DIY electronics. When not exploring the latest AI models, he is an avid maker and experimenter in the world of DIY tech.
Connect with Julian:
X / Twitter | Muck Rack
Building Your Own DIY Steam Machine is Easier Than You Think
Building a compact gaming PC capable of handling both 1440p and 4K gaming involves thoughtful planning and component selection, as outlined by SpecLog. Central to this project is the ASRock DeskMeet X600, a small-form-factor barebone system that includes a compact case, integrated motherboard and a … [Read more...] about Building Your Own DIY Steam Machine is Easier Than You Think
Complete Breakdown of the Gemini 3.5 Pro, Claude Lab, and Xiaomi MiMO 2.5 Updates
Google's Gemini 3.5 Pro and Xiaomi's MiMO 2.5 represent significant updates in AI technology, addressing both performance and accessibility. As noted by World of AI, Gemini 3.5 Pro introduces the "X-High" reasoning variant, which enhances the system's ability to tackle complex, multi-step problems … [Read more...] about Complete Breakdown of the Gemini 3.5 Pro, Claude Lab, and Xiaomi MiMO 2.5 Updates
Opus 4.8 Just Dropped: the Hidden Features You Need to Try Right Now
Opus 4.8 has arrived, bringing a host of updates to the Claude Code AI model that aim to refine its functionality and address prior limitations. Nate Herk explores how this version builds on Opus 4.7 by introducing features like dynamic workflows, which allow users to break down complex tasks into … [Read more...] about Opus 4.8 Just Dropped: the Hidden Features You Need to Try Right Now
Why Google DeepMind’s CEO Says True AGI is Still Decades Away
Google’s recent statement on Artificial General Intelligence (AGI), delivered by Demis Hassabis, CEO of Google DeepMind, has provided a detailed update on the challenges and progress in the field. Hassabis clarified that AGI remains a long-term objective, emphasizing the gap between current AI … [Read more...] about Why Google DeepMind’s CEO Says True AGI is Still Decades Away
Why Anthropic Released Claude Opus 4.8 Just 40 Days After Its Last Update
Claude Opus 4.8 introduces practical updates designed to address specific challenges in development workflows. Prompt Engineering highlights how this version incorporates dynamic workflows, allowing developers to automate tasks such as code migration and bug detection through parallel sub-agents. … [Read more...] about Why Anthropic Released Claude Opus 4.8 Just 40 Days After Its Last Update
How to Finally Fix Broken Formatting in Excel Dynamic Arrays
Dynamic arrays have brought significant flexibility to Excel, allowing formulas to automatically expand or contract across cells. However, this functionality introduces a common issue: formatting inconsistencies. For instance, when an array grows, newly populated cells often lack the desired … [Read more...] about How to Finally Fix Broken Formatting in Excel Dynamic Arrays
Valve Delays Steam Machine and Raises Steam Deck Prices
Valve's recent decision to significantly raise the prices of its Steam Deck OLED models has sparked widespread discussion among gamers and tech enthusiasts. As detailed by ETA Prime, the price hikes are substantial, with the 512GB model now costing $789, an increase of 43%—and the 1TB version … [Read more...] about Valve Delays Steam Machine and Raises Steam Deck Prices
Why Prompt Caching is the Secret to Slashing Your AI Costs By 90%
Prompt caching has become a vital strategy for managing the rising costs of large language model (LLM) operations. By reusing previously computed data, this approach minimizes redundant computations, significantly reducing both expenses and latency. Prompt Engineering highlights key techniques, such … [Read more...] about Why Prompt Caching is the Secret to Slashing Your AI Costs By 90%
3 Immediate Steps to Adapt to Google’s AI-First Search
Google’s recent overhaul of its search engine introduces an AI-first approach that prioritizes context-aware responses and zero-click searches, fundamentally changing how users interact with information online. Marketing Against the Grain examines this shift, highlighting how AI-generated results … [Read more...] about 3 Immediate Steps to Adapt to Google’s AI-First Search
$100 Steam Machine : How to Turn a Server Blade Into a Gaming Powerhouse
The ASRock BC250, a server blade repurposed for gaming, features a PlayStation 5-based APU with a Zen 2 CPU and RDNA2 GPU. According to ETA Prime, the GPU is configured with 24 compute units (CUs) by default, but users have found ways to unlock all 40 CUs, significantly enhancing its graphical … [Read more...] about $100 Steam Machine : How to Turn a Server Blade Into a Gaming Powerhouse
DeepSWE AI Coding Model Benchmark Finally Solves AI Training Data Contamination
DeepSWE, created by DataCurve offers a benchmark for assessing AI coding models by focusing on real-world programming challenges rather than synthetic test cases. According to Matthew Berman, one of its defining features is the use of contamination-free tasks, carefully curated to ensure models are … [Read more...] about DeepSWE AI Coding Model Benchmark Finally Solves AI Training Data Contamination
