Thomas Monk

Economics PhD, @LSEecon & @CEP_LSE. Labour and tech change.
Homepage:

[Thomas Monk @ttmonk]

Run (MacOS/Linux):
curl -LsSf https://t.co/RBh8VNlEhQ | bash

or (Windows):
irm https://t.co/vA0XbucNAm | iex
in your terminal today!

[Roberto-Rafael Maura-Rivero @rrmaura]

Can microeconomists help build better AI models?

New guide with @pavelkireyev_io: we map how microeconomic theory connects to the post-training of LLMs.

[Thomas Monk @ttmonk]

have found pi to be a really useful, minimal harness, working super well with @opencode go.

[Thomas Monk @ttmonk]

I've been working on pi-goal-x for exactly this use case, in https://t.co/KhkDEqKeXG: https://t.co/YuEemS77jn. using it on a daily basis for research

install with
curl -fsSL https://t.co/SsCrY7pQwK | sh
pi install npm:pi-goal-x

[Peter Steinberger 🦞 @steipete]

Here’s your monthly reminder that you shouldn’t be prompting coding agents anymore.

You should be designing loops that prompt your agents.

[Thomas Monk @ttmonk]

Run (MacOS/Linux):
curl -LsSf https://t.co/RBh8VNlEhQ | bash

or (Windows):
irm https://t.co/vA0XbucNAm | iex
in your terminal today!

[Thomas Monk @ttmonk]

super impressed by the team at @MATLAB!

My tools mcp-stata and Stata Workbench try to provide exactly this 'agentic toolkit' for Stata (within any IDE!). Take a look:
https://t.co/MJXpAfITxc
https://t.co/6wOfiXwgxu

[MATLAB @MATLAB]

Introducing the MATLAB Agentic Toolkit

Get started 👉 https://t.co/efiOXdSM6z

[Thomas Monk @ttmonk]

i'm biased (Claes recommends my Stata Workbench extension, https://t.co/6wOfiXwgxu) - but a super useful guide for economists getting into AI within workflows we already use

[Claes Bäckman @ClaesBackman]

I put together a short practical guide for economists who want to use Claude Code, but who haven't gotten around to trying yet.

The goal is to reduce the start-up costs by using Claude Code within VS Code.

[Arpit Gupta @arpitrage]

Proposal: grad students should get a budget for compute

[Thomas Monk @ttmonk]

@BeatrizGietner steve machin/sandra mcnally/steve pischke (and many others!) here at the CEP?

[Thomas Monk @ttmonk]

@caterina_vieira @IIES_Sthlm congrats again - such a great outcome!!

[Thomas Monk @ttmonk]

@BrendanNyhan completely agree that it's an option. i've moved from the command line to the mcp server. abstraction helps.
it allows for multiple sessions, easy data browsing, help files being handled properly, errors being caught etc. all of this is abstracted without requiring more token use

[Thomas Monk @ttmonk]

built a lightweight mcp server for stata. run commands, check variables, and debug code through claude/codex/antigravity/windsurf without copy-pasting stata output into the chat. ~700 lines of python

https://t.co/kZuoyrL2BD

[Thomas Monk @ttmonk]

@cblatts hi Chris, great to see others getting claude-pilled! just flagging some Stata tools I've made that might be of interest to you/others
https://t.co/gWuBMWFT3p

[Thomas Monk @ttmonk]

built a lightweight mcp server for stata. run commands, check variables, and debug code through claude/codex/antigravity/windsurf without copy-pasting stata output into the chat. ~700 lines of python

https://t.co/kZuoyrL2BD

[hillaryvipond @hillaryvipond]

I am offering mentorship for PhD students in economic history again this year.

My PhD benefitted from having a fantastic supervisor, whom I can never repay. He said, "pass it on". I'm not an established academic, but I know what being a PhD candidate is like. 1/2

[Thomas Monk @ttmonk]

had a great time here last year - well worth applying!

[Alex Imas @alexolegimas]

Are you a student interested in learning the latest AI and ML methods for social science research?

Apply to our AI & Economics Summer Institute. Held at @ChicagoBooth and co-organized with @ethayarajh @LindseyRRaymond @sanjog_misra @asheshrambachan @danielrock

[Thomas Monk @ttmonk]

@YujungHwang3 asking Claude to write tests against expected output, and stating directly that success depends on passing these tests helps hugely with this

[Thomas Monk @ttmonk]

@hanjliu i think Stata was inspired by you
https://t.co/1JkiiJH5cs

[Stata @Stata]

💕 Send these to someone you’d be null without. 💕

[Thomas Monk @ttmonk]

@LSEnews @LSEPublicPolicy credit to @hanjliu

[Thomas Monk @ttmonk]

my mpa metrics class is v cute. look at how much fixed effects are improving their lives. @LSEnews @LSEPublicPolicy

[Thomas Monk @ttmonk]

@aniketapanjwani the thinking is: is the cli interface designed by StataCorp the best way of interacting with Stata. if not, some light middleware can improve this

[Thomas Monk @ttmonk]

@aniketapanjwani i thought the same. but after setting up a workflow almost exactly like yours, the mcp just made more sense. multiple sessions, data browsing, help files being handled properly, etc. all of this is abstracted without requiring more token use by the llm