About GeniusTechLab

I'm Josh, the person behind GeniusTechLab. This site is an independent technology publication focused on practical, hands-on guides for builders, tinkerers, and IT professionals. No press releases repackaged as reviews. No "top 10" lists assembled from spec sheets. I write about hardware and software I've actually configured, deployed, broken, and fixed.

I'm based in Adelaide, South Australia. I've worked in IT for years, including IT roles at Spendless Shoes (an Australian retailer), hands-on security assessment work, and running Gigantech ICT — a managed services provider (MSP) serving small and medium businesses in Adelaide. On the infrastructure side, I build and maintain production systems daily: UniFi networks, Docker and Proxmox homelabs, Ollama and local LLM deployments, and crypto trading bots. The guides on this site come directly from that work, not from a content brief.

What I Cover

The site spans five core areas, all connected by a common thread: infrastructure you can build and maintain yourself.

• AI & Machine Learning — Local LLM deployment guides, hardware benchmarks for inference (GPUs, NPUs, mini PCs), AI agent setup tutorials, and practical comparisons of cloud vs. local AI. If it involves running models on hardware you control, I cover it. I run Ollama, vLLM, and llama.cpp on my own hardware, so the numbers in these guides come from real benchmark runs, not marketing claims.
• Cybersecurity — Hands-on security hardening for homelabs and small business networks, VPN configuration guides, vulnerability breakdowns, and hardware security (Ledger, Trezor) reviews. I test attack vectors in controlled environments before writing about mitigations. My security assessment background informs how I approach these guides — I'd rather show you the exploit path than hand you a checklist.
• Crypto & Trading Infrastructure — Bot development guides, wallet security best practices, exchange reviews, and trading infrastructure setup (TradingView, API integrations, automated strategies). Not financial advice — technical infrastructure for people who manage their own crypto operations. I build and run crypto trading bots in production, so the code in these guides is the same code I depend on.
• Homelab & Networking — Mini PC builds, NAS setups, UniFi networking guides, Docker deployment patterns, and rack-and-stack tutorials for home and small office environments. Everything from a $100 learning lab to a full 42U rack. I run a Proxmox cluster at home and deploy UniFi gear for Gigantech ICT clients, so these guides reflect real-world setups, not lab fantasies.
• Best-Of Buying Guides — Curated product recommendations across all the categories above. Every recommendation is based on hands-on testing, community feedback, and spec analysis. I link to Amazon and other retailers via affiliate links, but recommendations are editorial — I don't rank products higher because a seller pays more.

Who This Site Is For

GeniusTechLab is written for people who build things. That includes:

• Homelabbers running Proxmox clusters in their closet, experimenting with self-hosted services, and learning enterprise-grade infrastructure at home
• IT professionals who need practical, no-fluff guides for networking, security, and virtualization that go beyond vendor documentation
• Developers integrating AI into their applications — running local models, building agent pipelines, and choosing the right hardware for inference workloads
• Tech enthusiasts who want to understand how things work under the hood, not just which button to click

If you're looking for consumer gadget reviews or smartphone camera comparisons, this isn't the right site. If you want to know whether a Beelink EQ14 can run Llama 3.1 8B at usable speeds, you're in the right place.

My Philosophy: Tested, Not Theorized

Every guide on this site follows one principle: I test what I recommend. That means running benchmarks, measuring power consumption, configuring the software stack, and documenting what works and what doesn't. When I say a mini PC achieves 18 tokens per second on a specific model, that's a measured number from my test bench, not a spec sheet claim.

This approach means my guides take longer to publish than sites that rewrite press releases. A single hardware review involves unboxing, OS installation, benchmark suite runs, thermal testing under sustained load, and real-world usage over several days. I'd rather publish one accurate guide than five shallow ones.

I also document failures. If a product doesn't work as advertised, I say so. If a configuration guide leads to a dead end, I update it. The comments section and community feedback catch issues I miss, and I incorporate corrections promptly.

Expertise & Experience

I'm Josh, the person behind GeniusTechLab. I've worked in IT for years, across retail IT operations, security assessment, and managed services. I currently run Gigantech ICT, an MSP based in Adelaide, South Australia, where I build and maintain production infrastructure for real clients — UniFi networks, Docker and Proxmox deployments, Ollama and local LLM setups, and automated crypto trading bots. This isn't a content farm or a side project run by someone who read a few blog posts. It's a publication by a practitioner who writes about what I do daily.

Specific areas of hands-on expertise include:

• IT operations in retail environments — including years in IT roles at Spendless Shoes, an Australian retailer with a national store footprint
• Hands-on security assessment work — vulnerability identification, network hardening, and practical threat modeling for small business environments
• Running an MSP in Adelaide, SA — Gigantech ICT, delivering managed IT, networking, and infrastructure services to local businesses
• GPU inference benchmarking across NVIDIA, AMD, Intel, and Apple Silicon platforms
• Network architecture — UniFi, OPNsense, MikroTik, and enterprise switching/routing
• Container orchestration — Docker, Docker Compose, Proxmox LXC, and Kubernetes
• Self-hosted AI infrastructure — Ollama, vLLM, llama.cpp, text-generation-webui, and local model fine-tuning
• Crypto infrastructure — hardware wallet security, node operation, and automated trading bot deployment in production

Editorial Independence

Some links on this site are affiliate links, which means I may earn a commission if you purchase through them. This never affects my editorial recommendations. I recommend products based on merit, not commission rates. A product with a 10% commission that's mediocre won't appear on my best-of lists. A product with no affiliate program that's excellent will still get recommended. See my Affiliate Disclosure for full details.

I don't accept paid placements disguised as reviews. If a company sends me hardware for review, I disclose it. If I recommend a product, it's because I believe it's the right choice for the use case described.

Advertising

This site participates in the Google AdSense program. Ads are served by Google and may be personalised based on your browsing history. You can control personalised ads in your Google Ads Settings. Display advertising keeps the site free to read and covers hosting, hardware, and time costs. I don't control which specific ads appear, but Google's policies prohibit malware, deceptive ads, and inappropriate content.

Get in Touch

Questions, corrections, or partnership ideas? Visit my Contact page or check out my Terms of Service and Privacy Policy. I read every message and respond to legitimate inquiries within 2-3 business days.