Chat apps and spreadsheets
Run research agents without writing code.
The research modes in ChatGPT and Claude can read a stack of sources on an idea, summarize them, and argue the other side. This route needs only a chat app and a spreadsheet.
Invest better with AI, and keep every decision yours.
Skip the AI that promises to run your portfolio. Use it to research your own ideas the way an analyst would, and to tell you when one stops adding up.
Choose your starting pathA publication by Nathan Slaughter
What you’ll use AI for
- Ideate
- Find the numbers that move a theme.
- Turn a hunch into a call you can test.
- Research
- Send an AI agent to read the sources and cite them.
- Pull the economic data behind an idea.
- Get the segment numbers, management’s discussion, and footnotes out of a 10-K.
- Monitor
- Write down what would prove you wrong, and hear about it when it happens.
- Get a news summary that covers only your ideas.
- Decide
- See each thesis next to its latest numbers.
- Set a price for when you’re right and one for when you’re wrong.
What this looks like
AI found a risk hiding in two of my ideas.
I had separate ideas about memory supply and power. In one conversation, an AI pointed out that both might depend on the same thing: big cloud and computing companies keeping up their capital spending.
If that spending slows, both ideas could weaken together. It also suggested what to watch, and I used it to decide what to research next.
What AI changes
AI can do much of an analyst’s legwork.
Give it a theme, and it can break the theme into the numbers that drive it and pull the economic data that measures them. It can find what matters in a 10-K and argue against your thesis the way a skeptical investment committee would. Coding agents can automate the fetching, parsing, and summarizing that crowds out the thinking.
What you can do that a fund can’t. You don’t have a benchmark to track, a style box to stay in, or clients to explain a bad quarter to. You can buy companies too small for a large fund to bother with, and tie every holding to your own theses. The freedom pays off only when the ideas behind it hold up.
Where do you want to start?
- 01
Investment Framework
Decisions you can trace to your portfolio’s job are easier to review
Turn your view of the world into specific calls, and use AI to research each one.
Explore this path - 02
Asset Allocation
Which allocation fits what you need the money for?
Have AI lay out how much risk each mix carries before you choose one.
Explore this path - 03
Stock Selection
If you pick stocks, start from screens you’ve tested
Find candidates with screens that hold up, let AI read their filings, and set your price targets before you buy.
Explore this path - 04
Portfolio Automation
Automate one recurring task at a time
Have AI watch your theses, summarize the news that matters to them, and rebuild your reports on a schedule.
Explore this path
Work in chat, or build your own tools.
Coding agent
Build tools around your own process.
A coding agent such as Claude Code or Codex can build small tools that pull economic data, parse filings, and rebuild your reports on a schedule. You describe the job, and the agent writes the code.
A coding agent built my backtester around my methodology. A friend who manages money professionally, with a completely different strategy across dozens of accounts, had one build his own. It draws on his custom data feeds and shows him what he needs to see to stay in charge.
Available now: Write down your investment process · Portfolio weight calculator · Allocation explorer