Eric Ma's Blog

Dan Chen and I have a tradition: every year at SciPy, we find a corner of the venue, grab coffee, and nerd out about teaching, talking through how people learn, how to teach better, and what actually sticks.

This is a post I have been wanting to write for a while.

I've been sitting on a question since SciPy this year.

I've spent the year on both ends of agent-written pull requests.

If you've ever spent an afternoon clicking through dose-response curves, flagging outliers by hand, there's a way to make the model do that automatically.

I took part in Round 2 of the alphaXiv x marimo competition.

I just got back from SciPy 2026, my eleventh year at the conference.

One year ago at SciPy 2025 I wrote a post arguing that laboratory data should live in a single xarray Dataset, with sample IDs as the shared coordinate system.

I had Qwen3.6 running on Ollama, hosted on Modal, and it worked.

I gave a talk on 10 June at Data-Driven Pharma (East) 2026 where I built a Bayesian hierarchical model of protein melting points live, in front of an audience, in 27 minutes (including live questions).

I was debugging a transcript duplication bug in my voice-first gym coaching app.

You are deep in a feature branch.

Voice-first AI means voice is the primary way you interact with the model.

My friend Sean Law asked me this question on LinkedIn recently, essentially, "Do you have a definition of the term agent harness?"

In my previous post, I compared three agent harnesses across workspaces, notifications, automations, and open-source status.

I have used three coding agent harnesses extensively over the past few months: the Codex app, cmux, and Cursor.

My teammate Jackie Valeri and I recently co-taught a workshop at BioIT World 2026 on standardizing data science ways of working.

I have been thinking about what the core mission of a data scientist actually is in 2026, surrounded by AI coding assistants, LLM-powered applications, and the relentless buzz around adaptive software development.

This year I ran a small experiment at ODSC East 2026.

This blog post is for anyone who is experiencing, or has witnessed people experiencing, AI-related burnout.

Marimo Pair has been released!

Since the beginning of the year, as I've been really maxing out on agentic coding and trying to explore the patterns and figure out what's working and what's not, one particular thing has been sticking out: I'm paralleling so much of my work.

I want to tell you about canvas-chat, a project I built with heavy AI assistance.

The problem with lean times When the economy tightens, formal development opportunities are usually the first things to go.

I owe this term to my colleague Wenhao Liu.

Agent skills are great, but I've been thinking about this... skills alone aren't enough.

This past weekend, I found myself revisiting a blog post from PyMC Labs titled "Running PyMC in the Browser with PyScript".

Folks have asked me how I do personal knowledge management (PKM) at work.

Speed without control is just chaos.

Having tasted what agentic coding could look like for software development, I wanted to know what it would look like for data science - this meant training machine learning models and answering scientific questions.

Most of the conversation about AI coding models focuses on performance metrics.

In part 1, I covered AGENTS.md as repo memory.

In part 1, I focused on repo memory with AGENTS.md.

I want my coding agents to get better every week.

I ran into a frustrating bug this week in canvas-chat, my experimental canvas-based chat interface I built at the end of last year.

I've been mulling over this idea since last year January: A visual, nonlinear interface for LLM conversations—something like an infinite canvas where you could branch, merge, and see the shape of your thinking.

Tommy Tang asked me about my opinions on OpenCode, so here's what I've learned after spending significant time with OpenCode and Claude Opus 4.5.

I had a beautiful tmux status bar on my old laptop.

Last year, I challenged myself to write one blog post per week, and I hit 53 posts by the end of 2024.

I’ve been thinking a lot about career lately.

I've been using coding agents for a while now, and I've learned a few patterns that make the experience much more productive.

Within research organizations at most pharma and biotech companies, professionally-trained statisticians are often staffed at extremely low ratios relative to the number of lab scientists.

I used to assume that coding agents like Cursor, GitHub Copilot, and Claude Code only work within a single workspace.

I recently discovered PocketFlow, a framework for building LLM-enabled programs created by Zachary Huang.

Coding agents promise to unlock significant productivity gains by working autonomously in the background—gathering context, running tests, searching documentation, and making progress on tasks without constant human intervention.

If you're like me, you might find coding with AI assistants somewhat addictive.

I spent time digging through Anthropic's skills repository.

Like cars that lose value as soon as they roll off the lot, LLMs become outdated as soon as their training sets are fixed.

When Omar Khattabe presented DSPy 3.0 at PyData Boston Cambridge last week, I finally had the chance to dig into a framework that's been generating significant buzz in the LLM development community.

This past week, I went on a building spree, a part of my ongoing ultralearning practice, and built multiple projects using AI coding assistants.