Category: Daily Report

This has probably been the most eventful month that I’ve had for a long time.

Giving myself the tactical stress of a hard deadline has been fun and invigorating… albeit mixed with the occasional anxieties.

I had been stuck in neutral for quite a while with the handbrake on. The decision to just go and do it, is starting to have ripple effects in ways that I didn’t really imagine.

The one thing I’ve noticed is how insane the demand is for information on AI. The uptake of ChatGPT must be unrivalled in market history (beyond software) and it boggles the mind that effectively one product is so versatile that everyone is using it in their own way.

I’d like to know how much of the available information we have actually trained it on now. I imagine somewhere there is a copy of a model which has actually been trained on all the copyrighted material the AI companies can find. If you were Google or Microsoft with all their infinite data – everyones emails, work… all the software code, all the books … all the websites … they must have a copy that is actually trained on everything; just not released that to the public.

Then there’s doubtless the military and intelligence agencies which would have had this technology for way longer. How far are they down the road with this? Backdoor into Facebook, link it with all the other data they have, and suddenly you can mimic every person … you can synthesise insights over scale.

That’s not even beginning to mention the robots that are coming down the pipeline…

More to come over next few weeks…

23 days ago I decided to temporarily stop pursuing client work and restart some sort of entrepreneurial dream – building some of the ideas I’ve had for a while. They’ve been bugging me for a long time, and it’s now or never really.

I’ve never been able to hold down a job whilst attempting to build something else on the side. The job ends up distracting me and takes most of my energy. Whether its a job, freelance or contract, it always takes away the focus.

I would never encourage anyone to do what i’m doing. Life is far easier getting a developer job and enjoying life. If you don’t want a job, contracting is lucrative and flexible. And if you are super organised you can do really well freelancing.

So far 10% of the timeframe has already gone and the rest will go quickly. But at least I will have done something of note with my year. If I really, really can’t get something resembling a foundation up and running within 7 months then it’s sensible to take the easier and saner route.

But for the moment, need to banish those negative thought processes. I’m pretty certain if I just keep going in the right direction eventually I will spot the opportunity. From what i’ve seen far there is an immense amount of interest in AI, and whilst a lot of people know what they are doing with it, there’s a ton that don’t. There’s even more that aren’t looking at it yet.

It’s become cliche to say that AI will change the world but I’ll say my bit:

1. AI has unleashed a lot of creative energy in the development world and innovation is happening very, very quickly. It’s like the human hive mind has been prodded and it’s reacted positively aggressively to it.
2. People who focus on the shortcomings are forgetting to asses where it will be in five years.
3. It’s not intelligent, but it mimics intelligence, hence ‘artificial’ – it is ultimately a very good world documentation service … almost everything you need to know has been compressed down to a digital knowledge cube. It might not be 100% correct all the time, but it’s more than good enough right now which is why uptake is so relentless.
4. The everyday person can use it. There is no technical barrier now to someone being able to make all manner of digital media. And it’s remarkably cheap.
5. When you combine Language Models with OpenCV (computer vision) and then pump them into robots – which are also frighteningly speeding up in as a reality… you realise language models are just another piece of a much bigger puzzle.
6. Eventually people will just communicate with a LM which interfaces with some sort of interface that lets them build ‘apps’ on … but I do think apps will change very much in paradigms.
7. Companies who embrace AI to reduce costs and increase efficiency and effectiveness will see massive profits. But there will be losers.

Anyway thats enough for today.

Cheers

In previous startups I got too involved in one aspect of the business to the detriment of the other areas. It became unbalanced.

When I had a fully working admission booking system for the UK Theme Park industry, I focused completely on customer support and product development. I failed to look at the bigger picture. And I failed to think about sales and marketing.

So when the 2008 financial crisis began to squeeze the UK Theme Park ticket market, they got dissolved within a few years and I was left with no new customers for the booking system.

Likewise, when I was running a web design company in London, I failed to do proper talent scouting so I could build a team. So I just ended up burning myself out by trying to do it all.

In hindsight, in both instances, I was in a great position, but didn’t realise it. And that could have been fixed simply by making sure I put time into different areas of focuses.

It’s not rocket science, it’s just realising that you just need to repeat certain actions with consistency in order to build up the results that you want long term.

Generally speaking you have these aspects of your business:

• admin / finance / hr
• sales / lead generation / networking
• marketing
• research and development
• production
• people

These all depend on the overarching vision you have, but once you have that you can focus on those areas. Even if you just put in 20 minutes to each of these a day, that consistency builds up.

Starting Point – Getting the basics right

It’s really important to support the business side of your life, with a positive and healthy personal side. If you treat your body poorly for the next decade just to try and get to some hallowed gold at the end of it, you are potentially in for a shock. Not only do you miss the point of life, but poor health is more difficult to reverse than maintaining good health.

Last year I began implementing some basic life habits:

Step One – Keep Your Life Clean!

• Clean your home / office up
• Declutter & organise
• Process & organise paperwork
• Have a digital tidy
• House & garden maintenance

It might seem a bit ridiculous talking about these things on a Startup Blog, but the point is, these things cover the bases. If you can do fifteen minutes per cycle (choose … 24/48/72 hour) … and the amount of times I can remember feeling super overwhelmed by an untidy office and paperwork everywhere. Likewise, even now my laptop is becoming a bit of a mess and I need to streamline things.

Also if you ever watch Kitchen Nightmares you’ll know that most of the restaurants are absolutely appalling – they don’t get cleaned simply because the owners haven’t created consistent cleaning habits.

A recent job at a startup on my first day the desks were just caked with dust. It just needed a bit of attention. I’m not saying it will change your business into a super success, but keeping your life clean is far more preferable.

Step Two – Life Admin

• Personal admin & errands
• Personal finance
• Business admin & errands
• Business finance
• Investments and crypto

Doing these things regularly really does help keep on you top of things. The challenge is doing them consistently, but they are nice and simple tasks that you can apply your willpower and discipline to.

In my early years I left accounts to the last moment, but now I keep on top of things.

Anyway thats enough for today.

Today’s Work

A good day… I did some more R&D work with AgenticAI – the python library for building virtual agents. Quite interesting really. And I started building out some migrations on the laravel project. Whilst I will be using python a lot more, at the moment it still makes sense to make the most of Laravel’s very solid functionality and integrate that with a python API.

See Agentic work diary for today

Really good news on one of the projects but wont jump the gun there.

Since I’ve had a cold winter without adequate central heating, I took the opportunity to warm up and sat and meditated with breath-work in the sun again in the garden. Honestly, being your own boss does have its benefits.

There’s a lot on at the moment. But when there’s too much to consider, it’s not productive at all … and so you need a better solution.

Today was mostly conversations.

Are conversations useful?

If they lead to valuable action action then yes for sure.

But if you’ve watched Stutz on Netflix, you’ll know that real world conversations – and hence socialising – are also vital for your own wellbeing and mental health. It’s more helpful if the conversations are interesting and not poisonous, but any social contact is better than none, funnily enough.

If you are working on your own a lot as a founder, it’s so important to have social contact to ground yourself in reality. It’s very good for your brain.

The sad reality of future AI is that people will be more and more communicating with AI generated feedback. But it’s down to each one of us to make sure you keep on talking, especially technical founders who are introverted.

I was listening to someone the other day who said that whoever is alive now… we are the final era of humans who knew what life was like before AI (and robots) started to take over and reduce the cost of knowledge and content creation down to almost zero.

Interesting new world!

In other news, did some more R&D on web crawling. It turns out you can use the Crawl4AI python package in tandem with a language model and it will automatically run it through your prompt. I’ll do a video on it another time but for the moment here is my code. It basically will rewrite the BBC article as an excited Arsenal fan.

    
    # Example 2: Using Pruning filter
    url2 = "https://www.bbc.co.uk/sport/football/live/c8j00ke2r23t"
    success2, content2, file2 = await crawl_url(
        url=url2,
        filter_type="llm",
        llm_instruction="""
        Rewrite this as if you are an excited Arsenal fan.
        Include:
        - Emotive descriptive language of the goals
        Exclude:
        - Navigation elements
        - Sidebars
        - Footer content
        Format the output as clean markdown with proper paragraphs and headers.
        """,
    )

and this is the definition of my custom function

import asyncio
from crawl4ai import AsyncWebCrawler, BrowserConfig, CrawlerRunConfig, BM25ContentFilter, CacheMode, DefaultMarkdownGenerator, LLMContentFilter, PruningContentFilter
from crawl4ai.async_configs import BrowserConfig, CrawlerRunConfig, LlmConfig
import os
from dotenv import load_dotenv

async def crawl_url(url, filter_type="prune", query=None, llm_instruction=None):
    """
    Crawl a URL and apply a specified content filter.
    
    Args:
        url (str): The URL to crawl
        filter_type (str): Type of filter to use - "bm25", "prune", or "llm"
        query (str): Query for BM25 or Pruning filters
        llm_instruction (str): Instruction for LLM filter
        
    Returns:
        tuple: (success, markdown_content, output_filename)
    """
    # Load environment variables from .env file
    load_dotenv()
    OPENAI_API_KEY = os.environ.get("OPENAI_API_KEY")
    
    # Select the appropriate content filter based on filter_type
    if filter_type == "bm25" and query:
        content_filter = BM25ContentFilter(
            user_query=query,
            bm25_threshold=1.2,
            # use_stemming=True
        )
    elif filter_type == "prune" and query:
        content_filter = PruningContentFilter(
            user_query=query,
            threshold=0.5,
            threshold_type="fixed",  # or "dynamic"
            min_word_threshold=50
        )
    elif filter_type == "llm" and llm_instruction:
        content_filter = LLMContentFilter(
            llmConfig=LlmConfig(provider="openai/gpt-4o-mini", api_token=OPENAI_API_KEY),
            instruction=llm_instruction,
            chunk_token_threshold=4096,
            verbose=True
        )
    else:
        # Default to pruning filter with empty query if no valid filter specified
        content_filter = PruningContentFilter(
            user_query=query or "",
            threshold=0.5,
            threshold_type="fixed",
            min_word_threshold=50
        )
    
    md_generator = DefaultMarkdownGenerator(
        content_filter=content_filter,
        options={"ignore_links": True},
    )

    config = CrawlerRunConfig(markdown_generator=md_generator)

    async with AsyncWebCrawler() as crawler:
        result = await crawler.arun(url=url, config=config)

        if not result.success:
            print(f"Crawl failed: {result.error_message}")
            print(f"Status code: {result.status_code}")
            return False, None, None

        # Create a filename based on the URL
        # Remove protocol and replace special characters
        filename = url.replace("https://", "").replace("http://", "").replace("/", "_").rstrip("_")
        output_file = f"{filename}.md"
        
        # Write the extracted content to a markdown file
        with open(output_file, "w", encoding="utf-8") as f:
            f.write(result.markdown.fit_markdown)
        
        print(f"Content successfully exported to {output_file}")
        return True, result.markdown.fit_markdown, output_file