Posts

Showing posts from 2025

Remote Working Data Annotation options - Digital Nomad

Image
During the past, well, let's think, 10+ years, I've been involved in Machine Learning projects (okay, they call it "AI" now, but that's an argument for a different time) that require lots of data to be ingested / used to train the Machine Learning model. For instance, if it is text, we need lots of snippets of text that can then be categorised, so that the model "knows" that the 10 words grouped together in that context have a "meaning" and that meaning is labelled - this helps with questioning later on. It also helped to extract out the entities and relationships between the wording to give more context.  Here's a simple example using the spaCy tooling to give you an example: https://www.labellerr.com/blog/image-annotation-services-and-data-labeling-for-ai-models/ If it is imagery, we need lots of images with the segmented parts identified, usually a bounding box(?) to identify the elements / objects inside the image, so, again, this can b...

Nano Banana - quick go (scary!)

Image
Fancy trying out the new Nano Banana image manipulation tool? Well, originally I wasn't too bothered and then I tried it out with some images as starters and then got more creative with the prompts and was then actually very surprised at just how good it was at making fake imagery look really well done. You can literally just fake and make any imagery / scene that you can think of just from a base image. This is good, but, yes, can also be quite bad for obvious reasons - yes, there is a T&Cs violation restriction, however, it won't stop some weirdo getting hold of an online photo image of you and then including things in the image that were never there, like other people, etc... I'll show a few examples later, maybe. However, as this guy on YouTube shows, it can really be useful if you're working with your own imagery and want to be more creative: If you don't want to watch the video, I can give you a quick walk through of how to test out Nano Banana for free! Y...

GNSS / LoRA and Quantum Positioning systems?!

Image
This has been an interest and a concern of mine for quite sometime and it has been on my radar (pun intended) for quite some time to investigate some simpler or not so simpler alternatives. What's the problem?  well, GNSS (Global Navigation Satellite systems), otherwise known by the general public as GPS (Global Position systems) have a bit of a flaw in the fact that a particular country "owns" the satellites that are used, America owns one setup (GPS), Russian another (GLONASS), Europe another (Galileo), Chines another (BeiDou), India and Japan have their own Navigation Systems you get the jist. Interestingly it costs about $2Bn to keep GPS running and operational per year - however, it is "free" to be used by everyone. One of the BIG flaws of using GPS / guidance systems is the mass usage within drones - and the way that these signals can be blocked or overridden with false information - not a good thing :-( Now, imagine you wanted to "get off" the ...

Gemma 3:270

Image
  https://ollama.com/library/gemma3 ollama pull gemma3:270m compare the size to the other versions: here's a news article doing an analysis for you - saves me repeating it:  https://www.theregister.com/2025/08/15/little_llm_on_the_ram/ 270million parameters, but only taking up 550MB of RAM.  That is what makes this special.  On very low compute devices, Raspberry Pi, Jetson, low-spec laptops, this model can perform VERY well. I hear you say, "yes, but it might be performant and small, but it won't be able to do what I want it to do" Well, maybe you should go take a look at this article:  https://ai.google.dev/gemma/docs/core/huggingface_text_full_finetune What is that? it's the instructions on how to enhance / extend the model to be specifically knowledgeable about YOUR specific SME subject.   Yup, you can get all the specific knowledge that you want to train the model on & here are the instructions on how to then teach this tiny little pocket-rock...

Google Opal [experiment] - not in the UK yet

Image
Well, this is a turn up for the books and something long overdue for the general public: find out more here: https://opal.withgoogle.com/landing/ Now, you could just go off and do some gold-rush effort and I highly encourage you to do so - THIS is a game-changer.  It really is. As this is "my universe", I'm just going to pull it back to me (yes, very narcissistic, I know, but I don't care).  Here are a couple of articles that I just want to call attention to: https://tonyisageek.blogspot.com/2020/02/a-week-off-worksort-ofproject-o.html https://tonyisageek.blogspot.com/2020/12/project-o-my-machine-programming-project.html The basic premise was to use NLP (natural language processing) where you, as a non-technical person, could describe what you want an application to be and do.  Then, through the power of automation it would generate a starter-for-10 application from templated code and deploy it onto a Cloud environment, so that you could then see it / touch it / run i...

AI Agent software - n8n

Image
Yes.  I did just type that bunch of words.  I am so against using "AI" and the word "Agent", it actually makes me want to vomit.  It is buzzword bingo in its purity.  Getting beyond that, let me document a tool that could be really useful.  Potentially, I see the good old BPM (Business Process Management) software being used in conjunction with an orchestrated workflow process tool as being really useful for companies that want and need to do automation of business processes.  Not just jumping on the AI/LLM bandwagon. Let me introduce you to the community edition of: n8n So what is it? Well, it's kind of a bit like Node-Red in that it is a visual low code flow driven "programming" tool that allows you to hook up integrations really quickly and easily. As you can see in the screenshot above, "AI Agent" is the "thing" that has got people mostly interested.  What does that really mean? Well, it just gives a consistent repeatable workf...