How I talk to coding agents
The machine, the harnesses, the gated automations, and the one decision I make thirty times a day. Everything else is scaffolding.
Ran Aroussi (Official Website)
Software developer, financial tinkerer, and solo entrepreneur. Creator of yfinance and other popular Python libraries, founded Tradologics and offering CTO-as-a-Service at Automaze.
Latest posts
How I work
The machine, the harnesses, the gated automations, and the one decision I make thirty times a day. Everything else is scaffolding.
MUXI 1.0 Is Here
The server that runs intelligence. The prime time release - fourteen months from thesis to a cleared roadmap.
I'm Restarting the podcast
What I got wrong the first time — and what's different now
You need a Ghost in the Machine
How to use AI to fix the infrastructure blind spot – from the inside.
Self-Healing Software
Why the Architect model only works if your software can fix itself
The Case for Engineering Daemons
How an Architect running an AI-augmented pod actually delivers 10x. And why coding agents alone won't get you there.
The law firm model for software developers and agencies
Why the future of dev agencies looks more like Kirkland & Ellis than a SaaS startup.
You're using MCP Wrong. Here's what you should do instead
Hint: Ditching it for CLI isn't the answer.
I Let AI Analyze the X Algorithm Source Code
Spoiler: the code reveals exactly which actions matter.
How Anthropic is turning its strongest advocates into its loudest critics
This strategic signal risks turning Anthropic’s most passionate advocates into its loudest critics.
From Junior Dev to 10x Senior (How AI Can Accelerate Developer Growth)
How AI Can Accelerate Developer Growth (If You Use It Right)
I'm Writing a Book on Production-Grade Agentic AI (And You Can Read It Now)
There are three things that make AI systems truly "agentic". Most systems have zero.
Why "Agent Workflow" is an Oxymoron (And Why It Matters)
The fundamental flaw in how we're building AI systems - and what to do about it
The Illusion of Progress
Why the companies building faster horses will get left behind by the ones building cars
Spec-Driven AI Development
From writing code to architecting systems. Why developers should adopt the new development discipline for the age of AI
How we fixed the context problem in AI-driven development
Our project management system for Claude Code using GitHub Issues and Git worktrees for parallel agent execution
The funding gap: why revenue-based royalties are the missing piece for profitable builders
The missing funding framework for bootstrappers, indie hackers, and creators
Apple is spiraling... and it's showing
Why Apple's "research paper" is just a smokescreen
Vibe Brainstorming
How I test ideas, challenge assumptions, and break down complex problems
Agents Deserve Better than Retrofitted Frameworks
Why Building the Multi-Agent Future Requires Purpose-Built Infrastructure, Not LLM Wrappers
Old School; New Tech
Yes, I'm starting a podcast.
Building a lean AI-powered company
A vision for scaling without scaling headcount
The Rise of AX (AI experience)
The internet isn't just for people anymore
It's Your Fault
The hardest (and most liberating) truth
Get Rich Slow
The financial playbook I wish I had followed sooner
VIBE: Visual Interactive Build Experience
A white paper on a proposed methodology for software development that prioritizes architectural design and leverages AI-driven automation.
5 Niche business ideas you can start with zero money
I'll even be your first customer.
What would you do with 10x more agency?
With great power come great opportunities
The world's laziest trading strategy: Six years later
Do simplicity still wins? Plus, how a small change made a big impact
How I get things done: My productivity framework
Master your tasks and focus with my flexible, streamlined approach to productivity.
AI agents and the end of websites as we know them
As AI-powered assistants become the primary users of the web, traditional websites may soon be obsolete. Here’s why businesses, developers, and UX designers need to rethink everything.
Breaking the silence
Life has been a whirlwind. Here’s what I’ve been up to, why I haven’t been blogging, and what’s coming next.
How to optimize for luck
Hint: To win with luck, you need to know when the odds are in your favor
Fighting racism starts with a single baby step
While most people agree that black lives matter (why shouldn't they?), many also feel that the "PC Police" has gone too far with demands to ban movies like "Gone with the Wind", removing statues of historical figures, or renaming words like "blacklist".
Tradologics & The Future of Trading
An introduction to the Trading revolution: How to deploy a trading strategy with Tradologics - using Tradelets, Tradehooks, Pipelines, and the Universal API.
Algo Trading in the Cloud
What if there was a way for traders to **focus *only* on the trading logic** – without worrying about stuff like broker connectivity, data management, or infrastructure? Now there is...
It's time for a modern, standardized trading interface, suitable for the web-age
In this post, I share my vision for an Open Trading standard for communicating with online brokers using modern technologies.
Is this the world's best (and laziest) trading strategy?
Can you regularly beat the market, cut volatility by 40% and max drawdown by 70% with only four trades a year, or is this too good to be true?
Downloading option chain and fundamental using Python
The recently updated yfinance added a lot more capabilities to this already popular library. You can now download fundamental data , including company financials, balance sheet and cashflow, as well as option chain data . Here's how...
The Future of QTPyLib
I released the first version of QTPyLib , my Python library for algo traders, in 2016. If you had told me then that I would still be working on it three years later, I probably wouldn't have believed you. But guess what? That's precisely where I'm doing :)
Reliably download historical market data from with Python
Ever since Yahoo decommissioned their historical data API, Python developers looked for a reliable workaround. As a result, my library, yfinance , gained momentum and gets 300k+ 3M+ installs per month, acording to PyPi!
Machine Learning for Trading (with Python): Webinar Recording, Slides and Notebook (IV)
I had a great time presenting yesterday's webinar about Live Trading with Python ... This was the fourth and the final part of my webinar series on Treading With Python for futures.io 's members.
Fast Data Store for Pandas Time-Series Data using PyStore
As algorithmic traders, we need a lot of data to test and optimize our strategy ideas. Over time, the amount of data adds up and the search for a reliable, efficient and easy-to-use storage solution begins.
Volatility Index for Crypto Currency
The CBOE Volatility Index (Ticker: VIX ) is a well known measure of the stock market's expectation of volatility implied by S&P 500 index options.
My Bitcoin Scalping Strategy
Bitcoin has seen a tremendous growth in both price an trading volume this year. Noticing that, I’ve backtested a scalping strategy I normally use to trade interest rate futures on Bitcoin, and... it worked great with very minor modifications.
Live Trading with Python: Webinar Recording, Slides and Notebook (III)
I had a great time presenting yesterday's webinar about Live Trading with Python ... This was the third out of a four-part webinar series on Treading With Python for futures.io 's members.
Backtesting Trading Strategies with (pure) Python: Webinar Recording, Slides and Notebook (II)
On wednsday, I gave the second out of a four-part webinar series on Treading With Python for futures.io 's members. Here's the webinar's recording, slides, and Jupyter notebook.
You can still download data using Pandas Datareader using this quick hack
As you may have heard, Yahoo! finance has decommissioned its historical data API, causing many programs that relied on it to stop working. However...
Live Plotting in Python using Matplotlib and ZeroMQ
While working on a new dashboard for QTPyLib I needed to get a Matplotlib plot to refresh and update based on data coming from a ZeroMQ stream.