AI Flip Monkey

It all started back in October 2024 sitting in a court yard having a beer in Cusco Peru. Sat down with my mate discussing AI and all things tech. Off the back of this conversation and just when I needed it, vibe coding started to really gain traction and we decided to start building out our direct to consumer private blood testing company. I have been a BI consultant, Data Engineer, ML engineer and eventually into the messy world of corporate management. So while I had most of the skills required to build out our idea it would not of been possible to do it with my 9-5 without a lot of investment. Fast forward to June 25, I had a working product but it was vanilla. Functionally perfect, execution poor with no real brand or feel. We decided to work with a marketing company and the difference it made was enormous. Within no time we had a brand, a vision and a plan to get to market. And here we are today, the coming soon page ready with only one more bit of development left, early testing then time to launch. With it being such a slog, I’m glad to be at the end but hopefully the beginning of something. For all those who say just ship, i tend to agree. I have shipped a couple of apps, rebuilt my wife’s business CRM and open sourced a few bits all at the same time. But don’t feel pressured into releasing everything quickly. Some projects deserve the investment and the journey to get to the goal. Will it succeed? Who knows… But I bloody hope so 🤞

Built an LLM router in pure Rust. No GPU, no PyTorch, no preference data. 1.2M params, trains in 3.4 minutes on a laptop. 89.5% routing accuracy, 58% cost savings vs all-Opus. The trick: a sparse graph where routing nodes actually talk to each other. No existing router does this. Paper + code: <github.com/olliverc1985/A…> DOI: 10.5281/zenodo.19016401

I built an LLM router from scratch in pure Rust with no ML frameworks. Every existing router (RouteLLM, FrugalGPT, AutoMix) makes one decision: input → classifier → model. Done. AXIOM does something different — the routing nodes actually communicate with each other. The sparse computation graph has 4 traversal directions: → Forward (Surface → Reasoning → Deep) ↔ Lateral (same-tier nodes try before escalating) ↑ Feedback (Deep nudges Reasoning confidence upward) ⏱ Temporal (past decisions blend into current ones) No existing router in the literature has this topology. Results on 200 benchmark queries: • 100% accuracy on simple queries (zero false escalations) • 56.6% cost reduction vs routing everything to Opus • 735μs routing latency — under 1ms, no GPU • 1.2M parameters, 3.4min training on commodity hardware --- Surveyed 75+ routing systems for prior art. None use graph-based multi-directional inter-node communication. Built this as a solo independent researcher alongside a day job. Happy to answer questions.

The brain doesn’t receive reality. It predicts it, constantly running priors, correcting for error. Perception itself is predictive. Dismissing LLMs as “just next-token prediction” while ignoring that your own cognition runs on a strikingly similar principle. One evolved in carbon wetware. The other engineered on silicon. The algorithm doesn’t care about the chemistry. Prediction is the mechanism. Substrate is just the medium.

Playwright cli uses 94.3% fewer tokens for the exact same work. Full report + reproduction steps quoted below. If you’re running browser automation with MCP servers, you’re probably burning insane tokens on schema tax. Real benchmark shows @playwright wins

Now go and make me passive income with zero risk please.

Managed to miss them this morning, the last 10 minutes nothing is making it through. Enforced break it is then

Lots of 500 errors from Claude Code?

So Claude Codes remote control feature, I’m split on this.. already struggling to take breaks as I’m fully locked in, but when I do get out I need to disconnect and not be looking at a screen. Now we have a feature that allows us to continue from our phone. Starting to feel like an addiction that I can’t walk away from. Time for a dumb phone

@PeterMcCormack @RestoreBritain_ It’s exactly what the country needs, the issue is the legacy media will do everything they can to make everyone think they are in the minority. They can feel their days closing in fast

Rupert Lowe will become Prime Minister. The establishment can’t win against zero fucks leadership.

@darrengrimes Of course he is mentally unfit, his actions clearly reflect this. No reason for him to not stand trial and get banged up for life

So he gets a nice life in a secure hospital at our expense then? Deport him back there, he ought never have been here. Our suicidal empathy allows these men to stay and commit these atrocities. God rest you, Wayne. I am still so very sorry.

@Hesamation It’s awful, feels like a load of senior management (I can say that I’m one of them) who know zero about AI but using AI to write AI slop posts about fuck all

I just got off LinkedIn. my eyes are burning. it's a time machine to AI three years ago.

@DearS_o_n It’s fine stop worrying, couple of days of pain then it’s done. Most interesting thing, there a room in the hospital dedicated to knocking one out

MEN: have you had a vasectomy? And if so, what was your experience?

@The_GreatBonnie Or bring them into the chat as context

Unpopular Opinion: To save AI credits and time as a vibe coder, always check the docs first.

@DrPhiltill Sounds like the start of 2025 all over again. Hype hype hype, need to stay in a job. I’m not saying the models ain’t getting better but AGI is starting to sound a bit like the boy who cried wolf

Rumors I’m hearing from people working on frontier models is that AGI is later this year, while AI hard-takeoff is just 2-3 years away.

Ah ha, I have just the thing for you lol! I built pace coach as i always rushed when doing presentations. The whole idea behind it was to have a visual representation of my cadence when talking. Happy to drop you a promo code if you have an IPhone and want to download a copy for free. pacecoach.co.uk

Any other solo founders out there building alone? Marketing is by far the hardest part for me, what’s actually working for you right now and what’s not? Any advice is appreciated.

It was the horrible realisation I had when I started rolling out AI coding tools where I work. The seniors are knocking it out the park and the juniors are struggling to keep up. Given how much you can now do with seniors it’s hard to justify the juniors. while my conscious knows I need juniors for the future I will have moved on to somewhere else before it becomes an issue so I accepted it as it’s someone else’s problem

junior devs are fucked