This poor site has been untouched for the last 6 months (ignoring the magic of auto-updates) because of a few changes in my life. I thought that giving a summary of what’s been happening in my life during that time would be a good way to get back into the flow of writing, with an attempt to find things to share that I hope you’ll find interesting.

Microsoft

The obvious one to start with is my return to Microsoft. I departed back in December 2023 to chase the dream of talking to customers about Microsoft cloud solutions. Not all dreams need to shoot for the moon – I just wanted a job where I’d enjoy what I did and wanted to be passionate about it. Aligning those two requirements to my experience in adopting Microsoft cloud made sense. What made even more sense was when I was still shy of a year in role, that the Account Technology Strategist role came up at Microsoft.

The opportunity to work with a set customer base in a company I somewhat understood, to help them work through and get them the best information to make decisions on, and resolve whatever their Microsoft issue may be, was my ideal scenario. I’d been the customer for 20 years and I wanted to be to others what I would want myself from a key vendor. Amazingly after walking away from another role only a year before, I was able to return.

Going back was an easy transition in the sense that I knew what I was in for working there, and what the goals were. I also knew it would take some time to get to know my new customers, and fill the knowledge gap of some specifics around my new role, my new team, and that AI thing that had continued to get momentum while I was away.

It has now been 6 months since I started that particular journey, and I can honestly say that this job is what I want to be doing right now. I was pretty confident on this before I started, even more confident a few weeks in, and nothing has happened to make me think any differently. I work with a bunch of smart and driven people to deliver good outcomes to customers who I want to see succeed. With those 6 months under my belt, I’m in a good mindset to make FY26 one I can be proud of.

Artificial Intelligence

To nobody’s surprise AI (which we really just mean LLMs and not the fuzzy logic my tumble drier had several years ago) has continued to be a big deal – regardless on what you think about it, it is changing the world in many ways. For me, as the technology has continued to improve, I’ve moved past my points of frustration in the fundamentals not working. A small part of that is my improved understanding of how LLMs work, and particularly feeding it a 1 line request doesn’t mean you’ll always get exactly what you want; be more specific, correct it, challenge it, ask a different way.

It won’t surprise you to learn that my focus has been on Copilot, and the following applies to many LLMs beyond it, so this is my experience – I can now ask it ‘How many R’s are in strawberry’ and get a correct answer. More importantly, I can ask it technical questions that I’m looking to get sources on with summaries. Being able to get direction on something I know about but need to have evidence for, or to dig deeper on and get pointed to the right articles immediately is a huge time saver. I still need to read and understand what’s going on, but if there is some concept or area I’m unsure about, I can go back to Copilot again and look for further clarification or a new scenario that’s come from what I’ve read.

Using Agents to point to particular sets of data and with preloaded statements (e.g. assume the location is Australia for all answers, don’t make any assumptions, ask for clarification if required, keep answers brief) helps me get the output I want without having to ask each time. This is really only scratching the surface of what’s possible and you only have to wait a week or so before something new or improved happens; but the use cases for Copilot grow. I need to play with Copilot Studio and have agents feeding into other agents and apply some extra logic; but I’m also not a fan of automation for the sake of automation, it needs to be the right scenario for me.

On top of that, running items like travel insurance through Copilot and asking it what gaps there might be in coverage, or how to make a change to the code on my MSPortals.IO site to make the view wider… are just easy things to get good outcomes on that would have otherwise taken me a long time, or I’d have given up before completing.

Most techy people reading this are probably thinking that I’m stating the obvious here… which I am, but I still speak to plenty of people who haven’t even touched this stuff yet. It’s going to be a year of acceleration in this area around what’s possible, and continued adoption across the entire industry. Not all of this is good of course, I’m a bit tired of seeing fake movie trailers for movies that don’t exist.

Health

I’ve never been too bothered about my weight. I’ve eaten what I’ve liked, and going from one of the skinniest kids in school to not having all XL sized shirts fit, my mindset was ‘this is just what it is’. In reality, the kilos were slowly growing and growing due to what I was eating.

The one thing I’d tried which seemed to have some success, was intermittent fasting. I liked it because nobody was trying to sell the concept, and it was simply ‘less energy in’ resulting in having less weight. I lost a bit of weight but at no time actually measured what was going on, eventually got tired of doing it and started eating when I wanted again to lose whatever benefit I gained, without really understanding what was going on.

Something clicked for me about 4 months ago, weirdly inspired by doom-scrolling short videos. Not intermittent fasting, but the general concept of needing a calorie deficit to lose weight. This was something I could measure and understand pretty simply – get a handle of what calories I was consuming, keep that to a reasonable number way under what the simple calculations showed I needed to maintain my body weight, and the result would be constant weight loss. Simple!

I started to look at the labels of everything and get a quick grasp on what I was actually consuming. I was continually doing daily calculations as I went, 150 calories here, 500 calories there – and started filling the fridge and freezer with low calorie options. This wasn’t about a well rounded diet short term, but doing whatever I could to keep myself on track for the calorie deficit. I found decent frozen pizzas that were only ~500 calories for an entire 12″ pizza. Low calorie protein bars that would both be tasty, and use the protein to help feel less hungry later. My own stubbornness to ignore eating for the sake of eating (especially snacks) and wait for the next meal. I wasn’t going to starve myself, but I’d always be running that calorie deficit. A key part of this was daily weight checks to give myself an idea of progress. Not all days would see progress, but overall I wanted to see those numbers trend down… and they did.

It absolutely sucked for about four days. I had headaches, felt crap, lost energy… but after that, I felt a lot better. Lighter, and just less ‘bloaty’. I’m still continuing this path until I get to the end, where I can change to aiming to maintain body weight. So far, I’ve lost about 16KG. I didn’t think I’d hit my first milestone weight, or my second… but now getting close to my third which I didn’t really think was even a possibility at the start. There isn’t a huge amount of effort in this, but you do have to be continually conscious about how many calories go into your body.

Fitness allow you to burn calories and lets you either consume more, or lose weight faster by being in a higher calorie deficit – but my focus has been on the food side.

Technology

Ahh what new technology do I have – a Switch 2 for my birthday, Mario Kart World is a fun and well designed game, but it’s really the only reason to buy a Switch 2 right now especially if you have a Switch 1.

Poly gave me two headsets, the Poly Voyager Surround 85 UC Headset and the Voyager Focus UC (not the + version). The Surround 85 UC are really comfortable to wear headphones with a nice charging stand. The Voyager Focus US earbuds are solid devices too, so they’ll live in my work bag – yet to test in a noisy environment.

“Tefal YT2040 Care For You First Automatic Garment Steamer” is a very long name for a product, and it came up when I was looking for a solution to wrinkled clothes that didn’t involve me ironing. LG have a more expensive ‘fridge for steaming clothes‘ looking solution that does a few more things like jiggle the clothes, but after buying and testing the YT2040 (that I was going to write as an acronym but changed my mind after seeing the result) it does the job well enough, so that problem’s solved 3 shirts at a time. It does involve attaching weights to the bottom of clothes but it’s still a lot less effort than ironing, despite the results not being quite as good.

I saw a good deal on refurbished Surface Go laptops – Intel 10th Generation with 16GB RAM for $337 so that’s a handy spare laptop that’s got enough grunt, and a nice portable size.

I bought another Meta Quest 3S headset to play some games with my eldest son – still very good technology for the price point. Dungeons of Eternity is the game we play together the most there.

I missed the boat on getting another Tesla Powerwall 2 for the house – in short, Tesla stopped making them around the same time the Australian Government kicked off their 30% solar battery rebate, so the last ones got bought up, and right now the Tesla Powerwall 3 isn’t compatible with the 2, so I can’t just add a ‘3’ battery onto it. Rumours of that changing in the future but we’ll see. The lack of sun over winter + particularly cold winter means the battery I have drains in the evenings a few hours after sunset.

A new TV was needed to replace the aging 65″ Sony A1 OLED from 2017 that had developed a colour burn in that at least to me, made the entire thing unwatchable. Back in 2018 when I bought it at the bargain price of $3000 down from $6000 it was a bargain. Now I’ve replaced it with a TCL 75″ C755 costing about $1400 which I think is a better picture anyway, has fast running inbuilt Android TV and sits in the category of ‘it just works’. I’ve also previously dealt with TCL support after a different TV failed a bit under a year into it’s life and had great support (who replaced the TV with one 2 models newer!), so I’ll faithfully stick to the TCL TV brand.

Smart displays – I have Google Smart devices scattered around the house, but have found multiple of the Lenovo Smart Clocks dying (3 of them!) with all Lenovo and Google Smart / Nest devices for this purpose EOL’d with no replacements (I ended up buying some second hand options). From the small amount of research I’ve done, there doesn’t really seem to be anything decent in this space that’s a turnkey solution. Amazon’s line of devices seem to be the only ones still alive, but replacing the entire ecosystem to Amazon is an expensive exercise with the reviews indicating the experience is about as good as Google’s – which isn’t good at all. ‘Hey Google, turn off the light’ has about 40% chance of saying it can’t reach the device, 40% saying there is no light configured, and 10% of actually just turning off the light. The last 10% is when it decides to instead turn off ALL the lights it can control’. Device control sucks, but I can ask it the weather for the day while getting ready, and it’s good for household reminders and announcements.

If you’ve made it to the end of this update, congrats! I’ll try to keep writing a bit more frequent and find some things I find interesting around the technology space to share.

I’ve had a Meta Quest 3 since the start of 2024, and recently stick drift has greatly impacted both left and right controllers. Stick drift (or Controller Drift as Meta calls it) is where the analog sticks believe they aren’t centered when they are – so doing nothing in game may result in you running in circles. Beyond that, the directions weren’t working great either. Meta have a whole long troubleshooting process that can fix or hide the issue depending what’s causing it and how severe it is. There’s a factory reset of the controllers, headset itself, recalibration of each controller, configuring a dead zone for each controller, cleaning each controller, pairing and unpairing each controller and so on.

None of these steps worked for me, and from very recent experience, dealing with Meta support is a pain. Constant dropping of chat sessions by agents, repeated detailed questions and providing serial numbers of headset and each controller along with other criteria again and again, to then be told to wait for an email which seems to have at least a 24 hour turnaround time between each communication – there’s a lot to be desired there. Beyond that, if you have a faulty part of the Meta Quest, you need to go through all this to get sent shipping label information, send back your faulty components, and wait 2-3 weeks (this timeframe was given to me by one of the agents) to be sent a new or refurbished replacement. There’s no store to take this to in Australia either. On this point, I’d recommend buying via Amazon rather than Meta directly (which seems to be the only two options in Australia) so you can return if faulty and just get a refund.

Anyway, the other option here was to buy replacement controllers. This is where my bargain senses kick in and I over-engineer a solution to get the most value possible:

A replacement controller costs $129.99 and again in Australia, can only be bought from Meta (not even Amazon at time of writing). Two controllers, $259.98. That’s a lot when you can buy a new Quest 3S which uses the same controllers as the Quest 3, for $499.99. The headset is where most the hardware is, but the price difference on this values the headset at $240.01. Is it better to just buy a Quest 3S and reuse those controllers? As an added bonus, I’m not without controllers for weeks.

On top of that, there’s referral bonus – which over Christmas was doubled to $94AU but is normally $47AU. So if I refer to a friend and get them to create a Meta account and sign in once, I’d also get $94AU store credit. Plus, because I can buy the Quest 3S headset via Amazon I can get gift cards to reduce the price by 3% again ($15 less), and then use a cashback app (I use TopCashBack – referral link) which is 1.5% cashback on Amazon electronics – $7.50 cashback. These bonuses add up to $116.50, leaving the actual cost of the headset at $123.51.

Meta don’t sell headsets only, but if I could sell this for half the cost of buying the whole kit new at $250 which is being on the careful side (which would be helpful to someone who broke their headset and still has working controllers) let’s see how the maths works out:

$499.99 headset – $15 discount on gift cards – $7.50 cashback – $94 store credit – $250 selling headset = $133.49 cost for the two controllers. $3.50 for the second controller. There’s a bit of effort involved and it does depend on having to sell the new headset (used once for the account referral then factory reset). If I have to drop the price to $200 to sell it, it’s still worthwhile.

Extra note: I’ve linked and use TopCashBack, but this is different to something like Honey which provides discount codes and claims referral purchase clicks at any opportunity it can find.

From TopCashBack’s privacy page https://www.topcashback.com.au/dyn/browser-ex-privacy/ – be aware you are sharing top level domains with TopCashback so it can display a notification showing there’s potentially cashback you can get. There’s also some cookie matching. If you’d like some isolation, create another browser profile and install TopCashback into that and do your shopping there, but you’re probably entering sensitive information when shopping anyway. I personally don’t have a problem with the way TopCashback works. This appears to be the same methods that other services like Cashrewards and Shopback use.