What's the most ambitious project you have Fable working on right now?
Simon Willison, Django co-creator and Datasette creator, asks users to share the most ambitious projects they are building with Fable
The inquiry seeks qualitative examples of the tool's deployment
Many users praised Fable for efficiently handling ambitious coding tasks and producing non-slop output, while others complained about its high token costs quickly draining credits.
No Digg Deeper questions have been answered for this story yet.
Most Activity

@simonw open source codex level computer use for any agent https://github.com/minghinmatthewlam/computer-use-mcp

@simonw Nothing because I work in biology ☹️
What's the most ambitious project you have Fable working on right now?

@simonw It is building a 3D voxel-based solver for routing the 40+ cables for a parametrically specified digital twin of a local inference server we're designing, alongside designing the full topology of the server (8 RTX 6k PRO GPUs via a two PCIe switches). A lot of components.

@simonw I made this 3D video infographic explaining the AuthaGraph map projection, with population density/3D terrain and a moving day-night terminator. At least this felt ambitious until fable chewed through each of my requests like it was nothing...

I've been blogging a few of my Fable experiments today, most recently I had it more-or-less one-shot a CLI coding agent on top of my LLM Python library https://simonwillison.net/2026/Jul/2/llm-coding-agent/

@simonw Trying to get fable to work instead of routing to opus lol

@simonw TUI in Elixir isn't as easy as in Rust, so I'm having Fable see what we can do about that so development can remain agile.
https://github.com/bmanturner/foglet-bbs

@simonw Filling out PDFs of scanned pages using low because annoyingly Opus is still shit at it
It's such an annoyingly simple task I'm shocked it's so hard for other models to do it
No wonder Docusign makes money

@simonw Honestly, I gave Fable what I thought was ambitious. It one shot it in 1.5 hrs (12.5% of my Fable usage for the week, ~$50+ API equivalent cost). I'm now recalibrating what I think is "ambitious"... and about to test if Opus 4.8 would have done it too if I just asked.

@simonw I'm building a personal health wiki / analysis dashboard augmented by 6 years of daily FitBit health data + 9 years of logged strength workouts to optimize the hell out of my sleep & training routines. At 40, I think this one's more impactful than most of the "work" projects I do

@simonw Fixing all my broken opus code

@simonw Fable helped me take my TUI model architecture viewer tool to next level!

@simonw Testing out Blender skills

@simonw Using it to coach up Codex to computer use Unreal Engine more effectively. Trying to recreate my likeness in MetaHuman with 0 knowledge of how to do it. Codex struggled for a few days. Now with Fable 5 it’s making strides.
pretty fun experiment so far

@simonw Got fed up with QuickBooks. I'm just having Fable build it for me

@simonw Having it work on improving boot times for emberd, a self-hosted microVM based code execution sandbox for AI agents
http://github.com/hdprajwal/emberd

@simonw figuring out how to train a global umap for a given embedding model

@simonw The ambitious part is Fable giving me back the attention I used to spend nudging Opus all day.
I’m proud of this project, but I’m not sure I’d call the project itself ambitious
https://github.com/stbenjam/skillsaw

@simonw fable's the causal-reasoning agent in an 8-agent paper-trading loop i built. feed it a macro event, it chains two or three orders out to specific tickers, the other agents pull prices and stress the thesis. best part: watching it kill its own weak chains before it places a trade.