Every project I have built with AI, June 2024 to July 2026

AI can now do almost anything I ask it to. Whether what comes back is worth having is the part I am still pushing on.

I’m a principal engineer in AI, working in AI enablement. In 2024 AI finished my sentences while I wrote the code. By 2026 one paragraph from me set a team of AI agents running for nine hours.

How to read this page

Three things I should tell you

None of this was tested the way I test AI at work, with logging and grading, and no company tool touched it.

Only finished, public projects are on this page, which flatters me, and I would rather say that than hide it.

Every judgement is mine, written down afterwards from memory, with no scores and no visitor numbers behind any of it.

Open them yourself

Twenty-four projects in all, and nine of them are live pages you can open right now, built on an AI subscription anyone can buy.

Phase one

Models, and no machines

Jun 2024 to Sep 2025 · 16 months

I write every line of code myself and use AI as a tutor and an autocomplete. Nothing here runs on its own.

Leash Belka — small molecule–protein binding with GNNs

Jun 2024Leash Belka — small molecule–protein binding with GNNsPredicts which drug-like molecules stick to which proteins. That's normally lab work.Models checked against a real answer

The model learns from molecules drawn as dots and links, then trains on graphics chips rented by the hour.

My first attempt at using AI assistance for coding, and a really fun one. It was the Leash Belka Kaggle competition — predicting small molecule–protein binding. The goal was to take my understanding of graph neural networks, learn PyTorch Geometric, and apply it to the competition dataset. AI's role was narrow and honest: I used ChatGPT and Gemini to learn the framework, and autocomplete inside Visual Studio while I wrote. Not code generation. It took four and a half months, and it was a great learning experience.

What it taught me How to build and train these models properly. I rented time on powerful machines to train the final ones, and what I learned there I used again on two later projects.

What AI actually did AI carried the domain encoding well — getting raw molecular data into a form the model could eat was where it helped most. It could argue the architecture about half the way. Beyond that it did nothing: I hadn't started using models for debugging at this point, and nothing about evaluation or keeping the repo honest involved AI at all.

It ran longer than anything else here, and it is the project where AI did the least.

ToolsChatGPT (learning PyTorch Geometric), Gemini (learning PyTorch Geometric), VS Code AI autocomplete

Code on GitHub

Code onlyJan 2025

Running a whole product process through Cursor

I had an AI code editor plan, schedule and build an entire website: a scrolling chart of world economic growth.

An agent is an AI left to run a task; a skill is the written recipe it follows.

An early experiment with Cursor, before I found skills and agents. I was trying to give different roles access to specific subfolders, then have AI run the whole engineering process itself — write the project plan, cut its own Jira tickets, group them into epics, break those down into implementation detail. I wouldn't call it a success. I battled it constantly, and the tooling was too early to support what I was reaching for. But it gave me the grounding for how to visualize the framework. Most of my current projects run on what I worked out here: how to organize context inside a folder structure, and how to let an agent reach that context. It's also the project that drove me from Cursor to Claude. Progressive disclosure, skills, and sub-agents were what I'd been trying to build by hand.

What it taught me A lot, with no measurable outcome — which is its own kind of result. The framework survived; the project didn't.

I built it to be published and never put it online, so there is nothing to visit.

Web pages that explain a hard ideaToolsCursor, borrowed community .cursorrules

Code on GitHub

Replicating CoGraphNet from the paper

Feb 2025Replicating CoGraphNet from the paperRebuilds a published research method that sorts medical article summaries into disease categories by mapping how words and sentences connect.Models checked against a real answer

A fun project, and the first where I leaned on heavy AI-assisted code generation rather than just completion. I replicated the CoGraphNet paper successfully on the dataset it referenced. Then I tried to port it to another dataset and couldn't quite get that running. I'd still call it a success: I reproduced the paper's results using AI to generate the code. It also built directly on the molecule–protein graph work — setting up a graph neural network, training it, deploying it, running it on RunPod. AI helped enormously with the models, and just as much with preparing the datasets. This was also where I used it to understand RunPod itself. It helped me set up and train there, though it couldn't yet drive RunPod for me. That came later.

What it taught me That I could take technical papers, architectures, and images and get AI to convert them into code. That realization is what made every later replication project possible. Dataset preparation, not modeling, is where it saved me the most time.

ToolsChatGPT / OpenAI code assistance, Cursor

Code on GitHub

Gemma in the browser

Mar 2025Gemma in the browserLets any website run Google's Gemma chatbot model inside the visitor's own browser rather than on a server.Tools other programs plug intostill open

A browser has nothing like the power of a server behind it, so only the smallest models fit.

A very small project, and my first time using Google's MediaPipe library. I wanted to write a wrapper that loads Gemma into the browser and see what an LLM could actually do there. It worked — you can load a model in the browser and run inference on it. I didn't take it much further, because browser models at that point couldn't do a lot beyond intent recognition, and the context windows were small.

What it taught me That this is nascent rather than limited, and the potential is large. It matters for privacy, and it matters for whether language models can run on smaller devices. If they can, a lot opens up — think of a model running on a robot that cleans a room, where you can tell it to avoid an obstacle that drops in front of it. I don't know how far this has come since. It's worth another look.

ToolsCursor, Google MediaPipe

Code on GitHub

Food prediction from mass spectrometry data

Jul 2025Food prediction from mass spectrometry dataWorks out which food a lab sample came from by matching the chemicals in it against nutrition records.Models checked against a real answer

A friend at UCSD reached out to explore some data, and we worked on it together. I'd call it a success based on the outcomes from the model we built. It was interesting for two reasons. I got to apply what I'd learned about graph neural networks, and I used AI to understand mass spectrometry — a domain I didn't know. There was no agent-style ticket decomposition here. This was purely about building a model for food prediction and understanding the data. It was all code generation.

What it taught me That AI could get me productive in an unfamiliar scientific domain fast enough to be useful to someone who actually works in it.

ToolsCursor

Code on GitHub

GraphRAG, explained by scrolling

Jul 2025GraphRAG, explained by scrollingA web page that explains, as you scroll, how AI answers medical questions by searching a map of linked facts. That search method is GraphRAG.Web pages that explain a hard idealive site

I've always been a huge fan of data-driven journalism and the scrollytelling experience. This was my very first attempt at using Scrollama with AI to build something I could present. I took two papers I'd read and presented them as a scrollytelling piece. I'd call it a success. It's one of the first I did, and you can see that in the work itself. Since then I've done it many more times, and I've gotten much better at understanding how to get the best experiences out of AI. Simple, and it worked. I have no complaints about it. It served its purpose.

What it taught me How to build a scrollytelling piece with AI at all. Everything I've made in the format since started here.

ToolsCursor, D3, Scrollama

Open the live site Code on GitHub

GraphRAG knowledge behind an MCP server

Sep 2025GraphRAG knowledge behind an MCP serverA server that feeds an AI assistant, section by section, my notes on building searchable maps of linked facts. MCP is the socket an assistant plugs into.Tools other programs plug into

My first project using Claude, and the start of my experimentation with MCP. I wanted to take what I'd learned building GraphRAG on earlier projects and put it behind an MCP server — a service an AI agent can call to look things up, instead of me re-explaining what I know every time. The idea was that an agent could use the server to work with an end user and construct the right knowledge graph for their own domain. I finished it, and I started using it as an MCP server on other projects at work.

What it taught me How to build an MCP server, and that my own accumulated knowledge was worth packaging as something an agent could reach rather than something I re-explained each time.

ToolsClaude Code, Claude Desktop, MCP

Code on GitHub

Phase two

The pivot

Oct 2025 to Mar 2026 · 6 months

In October I start the folder of reusable agents I still use today. By December they are running a whole project without me.

The shared toolkit — my reusable agents and skills

Oct 2025The shared toolkit — my reusable agents and skillsI keep a library of written instructions, each teaching an AI assistant one job: forecast a result, read a bank statement.The shared toolkitstill open

This is the most important repository of everything, because it contains all the agents and skills I used for every project that came after it. Every single one of them uses it — the AI investment value analysis, the 3B forecast, the cricket model, the MedGemma challenge, the IPL forecast, the Allbirds analysis, the investing notebook, paper synthesis, GPU programming, IPL progress, understanding soccer.

What it taught me Building this was an exercise in abstraction — taking thinking from various domains and turning it into a series of agents and skills I could reuse. The superforecaster agent is the clearest case: the same one drives IPL prediction, the World Cup projects and investment analysis. It's taking ideas from engineering and applying them to thinking itself.

ToolsClaude Code

Code on GitHub

Working with agentic systems, as a scrollytelling primer

Nov 2025Working with agentic systems, as a scrollytelling primerI built a scrolling web page showing how to hand an AI a method to work through rather than a question. Scrollytelling means the graphics move as you scroll.Web pages that explain a hard idealive site

The second presentation I tried after the MedGraphRAG post. It's a scrollytelling presentation of my thoughts on how to work with agentic systems. Nothing major to say about it. It's very similar to the MedGraphRAG one, and it's a useful primer for anyone who wants to work with agents and understand why agents and skills matter.

ToolsClaude Code, D3, Scrollama, GSAP

Open the live site Code on GitHub

Encoding Tetlock into an agent

Nov 2025Encoding Tetlock into an agentI worked out the odds, step by step, that Mariah Carey's Christmas song hits three billion Spotify plays by Christmas 2026.An argument I worked out with AI

A base rate is how often something like this has happened before, which is where a forecast starts.

A really interesting one. I'd just read Phil Tetlock's book on superforecasting, and I was curious how well I could encode those rules into an AI agent. This is where my superforecaster agent was born. The project was my first attempt at using it not just to forecast, but to gather the prerequisite data — the base rates, the specifics — and apply the techniques from the book. It was a fun project.

What it taught me That a reasoning discipline out of a book could be encoded into an agent and then reused. The superforecaster agent came out of this and went on to drive the IPL predictions, the World Cup work and the investment analysis.

ToolsClaude Code

Code on GitHub

The first end-to-end agent project

Dec 2025The first end-to-end agent projectI built a model that predicts what happens on the next ball of a Twenty20 cricket match, from everything before it.Models checked against a real answerstill open

Training a model means running it over the same data again and again, each pass sharpening it.

I finished this one, and it's one I want to revisit at some point. It was the first project where everything, end to end, was done by AI agents: understanding the dataset, performing the data analysis with me as the driver but AI doing the work, designing the different kinds of ML models, setting up training on RunPod, pushing the data there, running the training, and monitoring Weights & Biases. Every single one of those was done by an agent. I didn't get the results I wanted out of the model itself. But the success was that I started to become confident letting AI monitor and report back to me. I'd work with it to set up a training run on RunPod, and it would watch — and when it noticed the training error wasn't improving after a certain number of epochs, it would kill the run and tell me. That was the biggest thing I got out of this project.

What it taught me That I could hand over the whole loop and trust the reporting. Killing a run that has stopped learning, and telling me why, is the point where monitoring stopped being my job.

ToolsClaude Code agents, RunPod, Weights & Biases

Code on GitHub

Learning Erlang and Elixir with a tutor

Jan 2026Learning Erlang and Elixir with a tutorI had AI write me a course in two programming languages, Erlang and Elixir — 26 lessons I could run and change.Had AI teach me a hard subject

A fun one, and similar to learnbiostats. What I was trying to do was interactive learning of Erlang and Elixir. I completed it end to end, got all the basics right, and learned the language well enough to get started programming in it. Another one where I used AI as a tutor. Through the exercise I also worked with AI on the original idea — building a multi-agent framework in Erlang — and learned that it didn't solve the kinds of problems I wanted to solve. I dropped that idea, but I came away with the language and the framework.

What it taught me Enough Erlang and Elixir to start writing it. And that the multi-agent framework I had in mind wasn't the right tool for my problems — which I found out by working it through with AI rather than by building it first.

ToolsClaude Code

Code on GitHub

Live siteFeb 2026

Biomarkers from a biopsy slide

Reads a photo of bowel-cancer tissue, predicts what an expensive lab test would say, and suggests drugs.

The slides come from a free public cancer archive, and the model reads each one as thousands of tiles.

An amazing project. I did it for the Kaggle competition, and everything was done end to end with my custom agents: brainstorming what to build, researching the domain itself — oncology, its pain points, and the technology there today — pulling the slide data from TCGA, building the model to train a multi-instance classifier, designing the UI and how to present the information, running MedGemma and the various open-source Google healthcare models on RunPod for specific tasks, and deploying the UI. The entire thing was managed end to end by AI, driven by me. This was the first project where I brought multiple agents together — for brainstorming, estimation, understanding a domain, design, deep learning geometry, deep learning architecture, and cognitive design. It was the first project where I really started leveraging the power of multi-agent co-human development.

What it taught me This is where I started really working with multiple specialist agents. They brought in knowledge from various domains — agents I'd built to learn with, or designed from books I'd read. That let me work on this project much faster than I could have without them, because the encoded specialist knowledge was already there. I could lean on it to understand things, build things and design things.

Jobs I handed to a team of AI agentsToolsClaude Code with custom agents, MedGemma, Google HAI-DEF models, RunPod

Open the live site Code on GitHub

Code onlyMar 2026

Forecasting IPL matches, with real money on it

Predicts who wins each Indian Premier League cricket match and bets real money when the odds look wrong. A contract pays $1 if it happens, so its price is the odds.

Prices swing all through a match, so the same contract can be cheap one minute and expensive the next.

The first project where I put my agents to use and traded real money on them. I designed a multi-agent team using agents from my Claude repository to analyse and forecast IPL matches — player performances, data, web search — and I let the agents learn from their own history, which was logged. I acted on the analysis that came through. I learned a lot, because I had never bought a Kalshi contract before this, and I had never bet in my life. It taught me how to understand sports markets on Kalshi, how to read odds, how to tell whether you actually have an edge, how much to stake when you do (Kelly fractions), and how to stay disciplined buying contracts against in-game price variance.

What it taught me The agents were never meant to teach me any of that. They were designed to do game-by-game analysis, and I learned through the process of using them. This is the first project that showed me AI can genuinely teach you to do something when you encode knowledge from your own domains and then carry it into a parallel one.

Jobs I handed to a team of AI agentsToolsClaude Code, multi-agent team built from my reusable agents, Kalshi

Code on GitHub

Phase three

Machines and orchestration

Apr 2026 to now · 4 months

I build no more models. One set of instructions now splits a job across many agents at once, then pulls their answers back together.

A ten-agent dialectical valuation

Apr 2026A ten-agent dialectical valuationTen AI analysts argue what Allbirds shares are worth after the shoe company suddenly turned itself into an AI company.Jobs I handed to a team of AI agentslive site

A fun project. I designed a series of agents based on Aswath Damodaran's online courses on corporate finance and valuation, and put them to work alongside a cognitive design architect. The goal was to see three things: how well AI could pull data, how well the agents could run a financial analysis from that data, and how well they could present it in a coherent format. The page itself is the visible byproduct — it's what lets you judge whether they managed it.

What it taught me This was the first instance where AI could pull data, use it to run an analysis, and then present that in a coherent form. The intermediate outputs were vastly different formats — one numerical, one long-form prose and reports — but the final output was pitched at precisely the right level for a human reviewer. That mattered to me, because AI output is genuinely hard to review with all the intermediate text it generates.

ToolsClaude Code, ten-agent dialectical team, cognitive design architect agent

Open the live site Code on GitHub

A macro analysis pipeline, fanned out

May 2026A macro analysis pipeline, fanned outTraces big economic trends down to the companies they affect, works out what thirty are worth, and ranks them.Jobs I handed to a team of AI agents

A really interesting one. This is my notebook for doing end-to-end macro analysis. It's the first project where I have a controller that splits the work across many AI agents running at once — the orchestrator prompt controls how information should flow through a graph of parallelised agents, when it should come in, and what output should go where. The goal is portfolio construction: an internal analysis of macroeconomic factors, which industries those factors impact, then the companies likely to be affected run through valuations, and a potential portfolio report out the other end. I use it as a guide. I've not bought the portfolio as it stands, but I use the report to pick companies, or ETFs that hold a subset of them.

What it taught me That you could split work across many agents, gather it all back, and run a real investment pipeline on it.

How I use it I use the report as a guide to pick companies, or ETFs holding a subset of them.

ToolsClaude Code, many agents running in parallel

Code on GitHub

Live siteMay 2026

A learning machine for crop genetics

A course that teaches the genetics of breeding better crops, run by AI tutors that quiz you and set exercises.

Genes come in versions; the course covers which versions get passed on, and how much traits vary.

This was a project to teach me biostatistics, and it actually worked really well. I did the first two modules and learned about crossover, recombination, alleles, and how to measure variance at a population level. The interactive nature of the course was teaching me a lot. I just didn't have the bandwidth to go through all five phases. The interface was the structure it generated — a modular, incremental, Socratic process. It worked because the AI could sense when I was forgetting something. I could go back to a previous topic to remember things, or forward to the right next module, which I can't do in a traditional course. It would create sample datasets or pull real ones for me to analyse.

What it taught me Real genetics, not just a system that could teach it: crossover, recombination, alleles, and how to measure variance at a population level. I cross-checked it later against Yale and Stanford course material and everything held up. The spaced repetition is why it stuck — it gave me exercises, asked me questions, and made me recap.

Had AI teach me a hard subjectToolsClaude Code, Opus 4.8 (1M context), custom biostat-* agents, Workflow scripts

Open the live site Code on GitHub

Turning a research question into a written digest

May 2026Turning a research question into a written digestTurns one research question into many searches, collects the newest scientific papers, reads them, and writes one combined summary.Jobs I handed to a team of AI agents

Similar to the investing notebook — I did both around the same time, and they use a lot of the same techniques. It takes a high-level query, something like "tell me what's happening in retrieval over the last six months." It does query expansion, fans out into parallel lanes for agents to pull specific papers against their expanded queries, summarises the main points, does an analytical reading pass, then fans back in and synthesises a report out of it.

What it taught me This is where I started splitting work across many agents and gathering it back, all the time.

ToolsClaude Code, many agents running in parallel

Code on GitHub

A week with GPU kernels

May 2026A week with GPU kernelsA team of AI agents teaches me to program GPUs, the chips that run AI, by walking me into deliberate mistakes.Had AI teach me a hard subject

Similar to learnbiostats and the Erlang project. I designed a similar learning system, and very quickly — within the first week — realised this is not for me. I was quite bored by what GPU programming actually involves. It was still helpful: it saved me from spending money or time on books and resources, and it gave me a good sense of what it's like to wet my feet.

What it taught me That a week with a learning system is a cheap way to find out a subject isn't for you.

ToolsClaude Code, multi-agent tutor

Code on GitHub

A schedule builder for the World's Fair

Jun 2026A schedule builder for the World's FairReads all 556 sessions at a four-day AI conference, asks a few either-or questions, then builds your day-by-day schedule.Jobs I handed to a team of AI agents

This was for the AI Engineer World's Fair 2026 in San Francisco. I used it to plan which talks and sessions I wanted to go to. Nothing major to see here — it's on the page mostly so people know I attended.

ToolsClaude Code, four-stage agent pipeline

Code on GitHub

Live siteJul 2026

Every Ball Ever

Draws all 316,199 balls ever bowled in India's pro cricket leagues as dots, and shows how the game changed since 2008.

Fable is an AI model, one of the Claude ones.

In a dynamic workflow the AI decides its own steps; it is told what is wanted, not how.

A really cool project, and where I started letting the AI plan its own work — instead of me setting out every step, and stepping out of the specifics. I forced myself to think in abstract terms about what a project would need. The goal was to pull all the data for every single IPL and get AI to analyse how the game has changed, and I wanted it visualised in a cool new format, because I'm a fan of scrollytelling. Data journalism taught me about unit visualisation with object constancy, which I'd never known before. It produced some extremely valuable insights about the IPL. I thought it was a great success, because it was the first time I could see AI take a generic high-level prompt and a set of questions and run with it. It's also where I started using the grill-me skill to ask me a ton of questions about what I actually wanted out of it. It ran end to end — a multi-hour run, more than nine hours.

What it taught me Unit visualisation with object constancy, which I'd never come across before. And that a high-level prompt plus the right questions was now enough — I didn't have to specify the orchestration.

Web pages that explain a hard ideaToolsClaude Code, Claude Fable, planning its own work

Open the live site Code on GitHub

The Two Meanings of 26¢

Jul 2026The Two Meanings of 26¢A page explaining how betting-market prices work, using a World Cup bet that cost 26 cents twice and meant opposite things.Web pages that explain a hard idealive site

A hypothesis is a hunch about what moves prices, and testing one means checking it against the data.

A similar project. The high-level goal was to pull all the Kalshi contracts and give me insights. This is what set me up to realise that AI has arrived. What I got out of these two projects was massive data analysis, hypothesis generation, hypothesis testing, visualisation, report generation, dashboards, scrollytelling — and all of the orchestration changing without me giving directives to the project.

What it taught me That I now need to use everything I've learned to start asking extremely pointed questions of AI, to see what the research gives me back. These two projects are laying the foundation for what comes next in my work.

ToolsClaude Code, Claude Fable, planning its own work

Open the live site Code on GitHub

What it costs to train a model, and who set the price of an ad

Jul 2026What it costs to train a model, and who set the price of an adAnswers two specific money questions: what it costs to train an AI model, and how the US advertising market decided its own prices.Web pages that explain a hard idealive site

One of the last projects here, and where I pushed on AI capability by giving it specific questions to answer rather than open-ended research like the IPL or the FIFA contracts. I combined two. The first was the chip and training-cost market: what it actually costs to train a model, broken into data-centre electricity, personnel, data quality, data-gathering fees, licensing fees and inference costs. The second was the advertising market: its history, who the buyers were, who the sellers were, how much money went through it, who decided the price, and how audiences were understood. To answer them I let the model go and find the sources itself, but across a very constrained search space, and I built a knowledge graph out of what it found. Then I allowed unconstrained explanation within that knowledge graph, and that is what generated the second article.

What it taught me The tool doesn't set the ceiling — you do. AI is as good as the best person using it and as bad as the worst. The strange part is that the same person can be at either end of that range on any given day. It comes down to how creatively you decide to use it.

ToolsClaude Code

Open the live site The Economics of Intelligence Who Holds the Yardstick Code on GitHub

Live siteJul 2026

Regulation Time

Retells the 2026 World Cup through the $12.8 billion people bet on it, drawing every $75,000 traded as one dot.

A market is well calibrated when its prices match how often those things actually happen.

This was the project where I leaned hardest on letting the AI plan its own work — and it ran end-to-end: pulled the tick-level data, analyzed it, generated its own hypotheses, tested them, and brought me plots showing which ones survived. Implementation, ownership, AI-driven analysis: all of it was great. My push is on value: out of everything it found, only three or four hypotheses made me think "I need to watch for this on Kalshi next time." The AI could do it. Whether what it found was worth having is the part I'm still pushing on — and that's what I'm aiming at in the next projects.

What it taught me Trust contracts priced $0–$0.30 only when volume is high — that's where the market is well calibrated; avoid low-volume, low-price contracts. And watch how a contract on one game moves the price of a team that isn't playing in it.

Web pages that explain a hard ideaToolsClaude Code, dynamic multi-agent workflows (Fable 5 + Opus 4.8), gated pipeline

Open the live site Code on GitHub

Check me without trusting me

What kept happening

Reaching a “no” cheaply

Three times AI got me to a no in days instead of months, and the months I did not spend are the whole result.

Mar 2025Gemma in the browser Jan 2026Learning Erlang and Elixir with a tutor May 2026A week with GPU kernels

AI teaching me a new subject

Six times AI got me working in a subject I knew nothing about, from mass spectrometry to crop genetics.

Jun 2024Leash Belka — small molecule–protein binding with GNNs Feb 2025Replicating CoGraphNet from the paper Jul 2025Food prediction from mass spectrometry data Jan 2026Learning Erlang and Elixir with a tutor May 2026A learning machine for crop genetics May 2026A week with GPU kernels

Pages that move as you scroll

I have published eight of these. Open the first and the newest, and judge for yourself whether I got better.

Jul 2025GraphRAG, explained by scrolling Nov 2025Working with agentic systems, as a scrollytelling primer Feb 2026Biomarkers from a biopsy slide Apr 2026A ten-agent dialectical valuation Jul 2026Every Ball Ever Jul 2026The Two Meanings of 26¢Jul 2026What it costs to train a model, and who set the price of an ad Jul 2026Regulation Time

AI can now do almost anything I ask it to. Whether what comes back is worth having is the part I am still pushing on.

Three things I should tell you

Twenty-four projects in all, and nine of them are live pages you can open right now, built on an AI subscription anyone can buy.

The timeline

Models, and no machines

Leash Belka — small molecule–protein binding with GNNs

Running a whole product process through Cursor

Replicating CoGraphNet from the paper

Gemma in the browser

Food prediction from mass spectrometry data

GraphRAG, explained by scrolling

GraphRAG knowledge behind an MCP server

The pivot

The shared toolkit — my reusable agents and skills

Working with agentic systems, as a scrollytelling primer

Encoding Tetlock into an agent

The first end-to-end agent project

Learning Erlang and Elixir with a tutor

Biomarkers from a biopsy slide

Forecasting IPL matches, with real money on it

Machines and orchestration

A ten-agent dialectical valuation

A macro analysis pipeline, fanned out

A learning machine for crop genetics

Turning a research question into a written digest

A week with GPU kernels

A schedule builder for the World's Fair

Every Ball Ever

The Two Meanings of 26¢

What it costs to train a model, and who set the price of an ad

Regulation Time

What kept happening

Reaching a “no” cheaply

AI teaching me a new subject

Pages that move as you scroll