Journey Of Launching An AI Startup. Win or Perish Challenge 2025!

Author: admin

  • Day 104 – How to decide when to call it quits on a project, or when & how to pivot?

    For the last few months I have been partially helping out an AI startup get a handle on its product. Their lead developer had left a year ago, leaving a fairly complex and over engineered codebase.

    I realised last night how much of my time, attention, and energy I had been putting into the project.

    Whenever I find myself waking up and my automatic thought process goes to a particular project, I regularly need to assess whether that project is worth of my ever decreasing time.

    The problem is – as with almost all software projects at the moment, is that unless you are able to integrate AI coding tools with it to develop new features faster, you are falling behind.

    The reality is that the combination of automation tools + ‘vibe coding’ + ‘senior engineer experience’ can outpace and rebuild most systems from scratch very quickly now. The cost of software development is now significantly less although the increase in technical debt from AI generated code will be substantially higher than normal.

    Sunk Cost Bias

    With the right attitude, you can take lessons from most experiences in life. When an individual or company has invested resources into a project for too long, there’s this thing called sunk cost bias…which is the psychological desire to keep on going on a path because you’ve journeyed so far.

  • Day 103 – Google’s Move Kills Small Independents & Keeping Going…

    We’re now witnessing another phase of the consolidation of the internet into the hands of a few very powerful corporations. The original premise of the internet is dying, but there are things that we can do.

    Today, I’m commenting on a really long but excellent article which you should probably read if you are interested in the future freedom of the internet.

    It’s a long read but it paints a stark reality of what the big tech companies are likely going to do.

    Google is Using AI to Censor Thousands of Independent Websites Like Mine (And to Control the Flow of Information Online)

    My personal summary of that article is that:

    • Google have annihilated small and independent website owners over the last two years. Most have had to downgrade their teams, or shut up shop. If you look at the traffic statistics in that article, the drop-offs are terrifying.
    • The new move by Google is going to ensure that users get given the answers without ever leaving Google. They are happy to use other peoples content, but without rewarding it now.
    • It’s a complete power move that will see a handful of websites run by a very specific group of VCs – be visible in the Google’s AI results.

    I do think it will leave an area of opportunity for those interested in promoting the independents. I personally think Google has shot itself in the foot. It will still do really well because of its position in the market; but results have got worse and just more generic. A huge amount of the web is now hidden simply because Google deems it so. That will create an opportunity.

    I was speaking to a friend the other day and they asked where my startup updates were! Whilst I hadn’t been getting many people interacting with them, there was some awareness… and the comment made me realise that regular updates are the best form of ‘getting word out’ for when I have major milestones to demonstrate. So, I’ll be endeavouring to continue.

  • Day 102 – OpenAI Structured Outputs

    When you want the OpenAI API to return output in a specific structure, you use Structured Outputs.

    Here’s an example:

    import { NextResponse } from 'next/server';
import OpenAI from 'openai';
import { zodTextFormat } from "openai/helpers/zod";
import { z } from "zod";

export async function POST(request) {
  try {
    const client = new OpenAI({
      apiKey: process.env.OPENAI_API_KEY,
      organization: process.env.OPENAI_ORG_ID,
    });

    const requestData = await request.json();
    const { model, input } = requestData;

    const CalendarEvent = z.object({
      name: z.string(),
      date: z.string(),
      participants: z.array(z.string()),
    });

    const response = await client.responses.parse({
      model: model || "gpt-4o-2024-08-06",
      input: input,
      text: {
        format: zodTextFormat(CalendarEvent, "event"),
      },
    });

    return NextResponse.json({ 
      success: true, 
      event: response.output_parsed 
    });
  } catch (error) {
    console.error('OpenAI API Error:', error);
    return NextResponse.json(
      { success: false, error: error.message },
      { status: 500 }
    );
  }
}

    So, if you are doing it the javascript way:

    • You use Zod to define the object. Here you see we’ve created CalendarEvent with name, date and participants, all of which are strings.
    • You use the responses API and you use the { text.format }

  • Day 101 – Long term consequences of AI

    Day 101 – Long term consequences of AI

    It’s been quite a journey over last 100 days.

    I am very happy I made the decision to go all in on AI.

    I know a huge amount more than I did 100 days ago.

    I can see that it’s going to become a huge opportunity for big corporates to automate almost all their tasks.

    Big insurance companies will use AI to make most of their workforce redundant. Whether they find new work for them to do, I don’t know… but I can clearly see that the big companies are going to want to use AI to cut costs.

    Here’s where a potential down spiral begins:

    1 – Companies have always been interested in cost savings.

    2 – AI will offer the perception of massive cost savings.

    3 – Companies will either: a) ignore & wait b) upskill & create an internal AI team c) get in external contractors

    4 – Current internal systems and workflows will be analysed

    5 – Patterns in these systems / workflows will be identified

    6 – These patterns will be automated with AI

    7 – Companies achieve cost savings

    Obviously this is fairly abstract, but it’s sufficient to demonstrate what I’m talking about.

    So, all the big companies in the world will start investing in internal or external workforces (creating a new career sector) and they will reduce the bloat that humans created. So they will eventually, mostly, get rid of staff.

    How quickly and at what magnitude this happens I don’t know, but basically, take the UK as an example, all FTSE100 companies will eventually downsize significantly. This is simply because most ‘work’ that is done in enterprise companies is the moving around of data in a certain way. And in reality, most workers are a bit bone idle. There is huge room for improvement.

    As of January–March 2025, the UK’s labour force—comprising 33.98 million employed and 1.61 million unemployed, for a total of about 35.6 million people House of Commons Library

    Total employed (Jan–Mar 2025): 33.98 million people

    FTSE 100 companies: ≈ 6.0% of the workforce (≈ 2.04 million employees)

    Public sector (government): ≈ 17.4% of the workforce (≈ 5.90 million employees, excluding major reclassifications)

    So, let’s take a cautious approach – 10% of the workforce will be made redundant. So for every 100 people at AVIVA (insurance company), AI will absorb about 10 peoples jobs. That doesn’t seem unrealistic right?

    Let’s extrapolate to the entire 34 million. So 10% made redundant … thats 3.4 million jobs gone and not coming back … which triples the number of unemployed straight off.

    This is potentially manageable, but will put huge strain on welfare. Massive growth in government sponsored training companies as they try to achieve any sort of retraining numbers.

    The circular problem is:

    • 3.4 million people (many with kids) … suddenly have less money.
    • Less money goes into economy
    • Businesses shrink again, lay off more staff
    • Process repeats

    And the problem mainly is that the new brains are looking to create companies that are entirely AI driven. The fewer humans the better is the mantra. So you have that, and probably the fact that most companies will self-preserve and take out more than 10% of their workforce.

    The only way out of this is either:

    • Reduce supply chain costs (mainly through energy sources)
    • Open up new industries
    • Nationalise and subsidise industries
    • UBI

    To be continued.

  • Day 100

    I’m using this word press blog to get my original thoughts out.

    In a sea of AI generated content, originality is rare.

    So I’m going to carry on writing as I want to.

    That’s not to say that I won’t use AI generation, but I’ll do it in other areas. For a personal founder blog, generating anything with AI seems completely counter intuitive.

    Hitting day 100 was somewhat of a decent milestone.

    I will review over the next few days and weeks but I wanted to give some space to an excellent hour long video with a guy interviewing Gary Vee.

    The only thing I would say about Vee, is he has done excellent *but* he did build upon his father’s liquor store business. It’s always easier to take something and optimise it than build from scratch. So always bear that in mind if you are looking to compare your version of success to his.

    That aside, he’s so excellent to listen to.

  • Day 99 – Flowgramming with N8N is better than vibe coding with Cursor and most software developers don’t see the threat

    Everyone is going on about vibe-coding at the moment but the dark horse that is gaining traction, which I think is more likely to have a much bigger impact … are flowgramming tools like N8N, Make and Zapier.

    Whilst the automation tools are technically not AI, and Zapier has been around for ages anyway, there is a surge in creative usage of the tools – especially since you can easily automate with AI APIs.

    Key value proposition:

    • The cost reduction of creating useful outcomes is massive with automation tools.
    • They are easy to use and only require programming knowledge in somewhat advanced situations.
    • They are very visual, meaning you can retain a birds eye view of what you are trying to do, without getting lost in code
    • Most software is just a collection of functions in a workflow and automation tools provide the framework that lets you easily work on these functions

    Of course not all software will be replaced, but I have a friend who managed to create a automated voice assistant who takes phone calls, all with N8N. It took him several weeks of long hard days, but he managed it all himself without any programming experience.

    Developers who have software experience will always be at advantage in some situations, and of course as the number of juniors going into the industry die, old school programmers will be in demand to fix older software … so hang in there developers, there will be an upsurge in demand eventually.

    For myself, as I explore the automation tools, I now have to face reality and ask whether every business case I get asked to make for people, could actually be done within the flowgramming paradigm … and a large proportion can be!

    I honestly think businesses who don’t create automation networks within the next few years will start falling behind. It’ll take effort to work out what can be automated within a company but there are always manual tasks that take up time.

    Once you’ve skilled-up someone within the company on a tool like Make, N8N or Zapier … and they know how the company works … you can start picking off small tasks that consume time. If you can completely eliminate pesky tasks that take a few minutes every day for everyone in the workforce, or even attack larger problems … you start saving hours every week.

    For instance, emails are getting unwieldy for most people around me. I’ve never liked emails much anyway, although we are very fortunate to have an open protocol that allows us to communicate with one another.

    How many minutes a day do I waste opening up my email, scanning through them to identify urgency, or longer term relevance, deleting the spam … and maybe even making a quick response to a simple question… this is stuff that can now be automated by an pervasive AI that has context on your situation and goals.

    An example with N8N that takes an RSS feed and outputs an image for each item

    Results

    For the moment I just selected the first RSS item as an example, and then pumped it through to the OpenAI image generator.

    If you aren’t impressed, then you aren’t paying attention.

  • Automation Flows In Companies- Days 96, 97 and 98

    Over the weekend I had a ten mile walk into the countryside, and a four hour brainstorm with my business partner. All good stuff.

    Today I go back to N8N, the workflow solution. Am following along the official tutorial for the moment. It’s clear to me that all companies that embrace automation properly from the ground-up will save a huge amount of time, freeing up their staff to focus on the important stuff. Whilst not technically AI, automation is clearly part of the new paradigm change that will begin happening across enterprises, especially the SME sector which hasn’t really normalised it yet.

    N8N Scheduling

    If you are going to build an automated network of actions for your company, then having scheduled triggers are a must. Of course you can use CRON or jobs in frameworks to do this, but having it in a nice easy UI is just way better.

    In this example, you can see that you can easily setup these triggers with a lot of flexibility. They can be triggered every second or just once a month. For each time period, there are different types of options. So just having these triggers can give you some food for thought in what you might want to start automating.

    Off the top of my head:

    • At the start of every week I might want a script that goes out and captures all the relevant news for me from several RSS feeds. Those can then be put into a workflow.
    • Or I might want to send out an email to the managers of departments to ask for a progress report.
    • Or, could be an email shot, a sales report generation, a slack message, a LinkedIn message or it could check whether tasks have been done last week, and to alert whether there is anything top priority outstanding.

    Anyway, point is, that time triggers give you a huge amount of flexibility. In the above image you can see that these triggers begin flows (currently empty!) that start at the beginning and end of the work week.

    The NASA Node

    Whatever you might think about NASA and the space industry in general, it’s pretty cool that you can just access NASA information from N8N. It takes one minute to get an API key from them, you throw it in N8N and you can now get Solar Flare reports from the last week …

    You can also Test the step and see the output:

    N8N Conditionals

    You can then take the output from any node. In our case the NASA feed. What I like about this, is that N8N will show you the output of any API calls, and allow you to drag any field easily into the conditional statement. So here I have dragged ‘classType’ from the left and put it into the conditional, and searched for the string ‘X’ which is a class of solar flare.

    When you test, you get a lovely UI showing the true or false.

    The next step was outputting to postbin, which wasnt working as expected. I will continue with this tommorrow.

    Automation Software Cuts Costs

    I don’t know why I haven’t played with automation software more over the years. Probably because I was focused on coding. N8N abstracts away so much code to make this work, and unless you have a working foundation that can call any new API robustly and output it all easily … you would have to start from scratch and it would take a while. I just love how automation software takes away this hassle, and you can still integrate it with your internal systems if you want to.

    Licencing

    N8N has an enterprise licence agreement so if you ever want to self-host commercially in production, you still need to pay them a licence agreement. I totally get why they do this, but it’s annoying since I wanted to use this open source software freely. I think it means you can use it internally for free, but as soon as you want to build a product with it, the enterprise licence requirement kicks in.

  • Day 95 – HeyGen!

    Had a call with the guys yesterday at HeyGen.com

    https://www.heygen.com

    So these guys generate avatars for use in influencer marketing. Some key notes are:

    • They have an API you can use
    • Multilingual
    • Pre-generated ones can basically say whatever you want
    • You can speak to one of these avatars and they will talk back to you. Not pre-generated.

    During the meeting they were also able to take a 4 second clip of me in the video, and basically render it and make me say anything. The mouth movements and voice are not super great, but for a four second sample it was good.

    It’s clear to see we will just be interacting with these agents on websites, and as you talk to them, they will present the data necessary to complete your goal next to them.

    I imagine OperatingSystems will just have these inbuilt. Back to ‘Clippy’!

    High quality human avatars driven in real time by LLMs that you can talk to, will be the future of e-commerce.

    It was always going to happen, and it’s been tried before but now it looks like the tech is going to converge.

    Websites will just have a shopping assistant now that talks to you, suggests products, and you wont have to do as much searching, or even typing (which might be a good thing for those of us who’ve coded for years) …

    Psychologically humans will connect more with another human looking avatar helping them to complete their goals.

    You will talk to it, the LLM runs off using RAG to the objects available for sale, and then will come back the solution … and present it to the user. If it’s clothing, the avatar may just change clothing to demonstrate it.

    Potentially further into the future, the avatar will interact with the product, like holding a laptop, etc … unsure exactly how, but I imagine eventually all products will be showcased in 3D so the avatar will be able to work with that somehow.

    Interesting times.

    Every day I see something that makes me both sad and excited, depending on what perception I want to take!

    Vercel AI Chatbot

    Carrying on from yesterday, I managed to switch the local Chatbot to OpenAI and so now I have got the local UI up and running and hooked up to OpenAIs API. And I can change the colour scheme now.

    Ok, that’s it for today’s R&D update.

    Will also have some news on the DXP project soon.

    We’ll also go through N8N workflows from WordPress to the socials.

  • Day 94 – Every company will just have their own internal chat

    A quick detour away from N8N today in my R&D time…

    The time has come where the app I’m working on really needs a traditional ChatGPT interface.

    I did have a developer spasm and think about building the UI from scratch for fun, but …

    Fortunately Vercel do some amazing templates out of the box for different purposes:

    https://vercel.com/templates/

    Vercel do an official ChatBot one and there’s a guy on Github who does a more feature full one albeit hasnt been updated for a while

    https://github.com/mckaywrigley/chatbot-ui

    The official Vercel one is updated more regularly, so I’m just going to use that. Below is how it looks on the demo. I’m currently trying to configure some bugs in it, but it will be great to get this up and running locally so I can start testing it to see if I really do need to start from scratch. It’s already integrated with Auth.js which I dont really want.

    The other one is based on Supabase, which is an incredible offering. But will stick with this for now.

  • Day 93 – Nocode Automation Tools

    Yesterday took a good step forward on a stealth AI project I’ve been involved with since January. It’s operating in a crowded market, but we’ve recently identified a better way to pivot. Will be able to announce it soon, once final alpha tests have been done.

    Also this week we have potentially found a new addition to our management team – a proven AI lead generation specialist. But more on that another time.

    Today I began taking another look at N8N, I’d already tried a first time and was impressed. I was able to quickly create a workflow that read a webpage, summarised it, and then pumped it to Slack as a message.

    Honestly, these automation tools WILL be part of every single company going forward. They will replace a lot of internal software platforms.

    My first aim is to go from a WordPress post to publish on:

    • X
    • LinkedIn
    • Facebook
    • Instagram

    There is already a template setup, which is what’s so powerful about this N8N stuff – most of what you want to automate has already been done.

    • Installing N8N with Docker is super straightforward
    • Look at the automation templates: https://n8n.io/workflows/
    • For local use it’s super easy to create a webhook tunnel

    https://n8n.io/workflows/3086-publish-wordpress-posts-to-social-media-x-facebook-linkedin-instagram-with-ai

    Today’s video analysis is of the guy who made Windsurf – which I haven’t used yet, but am going to give a proper try. Might prefer it to Cursor!

    Main takeaway is:

    • Sales team combines AE and forward deployed technicals

    In other news, this is a real time object identification locally that I want to try out

    https://github.com/ngxson/smolvlm-realtime-webcam