Research Software Architect for Research Software Engineering at the University of Manchester.
Tcl programmer, but also many other techs too. Lots of software automation engineering these days.
I want real AGI, not fake stuff powered by overtrained LLMs.
| Best Language | Tcl |
| Best Project | SpiNNaker |
I live in a country (Ro) that went from a horrific Stalinist dictatorship to being a self titled 'original democracy' to being a full EU member state, and all these in only one generation.
And now I realize the more democratic the country became, the safer protests became.
From machine gun shooting, to isolated incident, to being safe to take a kid to a protest.
They perform projections in a (very!) high-order smooth differentiable space. For problems which that describes, they work excellently. But when the problem doesn't fit that pattern, they're not so great. (The underlying math of hypersurfaces doesn't require differentiability, but the training algorithm guarantees it.)
Public LLMs have an awfully large training dataset. Much bigger than you think.
I've just finished making as many bits and pieces of TclOO be implemented in C as I can, both for 9.0 and 9.1. (No, I'm not backporting this to 8.6.)
Total acceleration of the interpreter create/delete cycle: 33%. That'll do for a couple of weeks' work. But let me know if I've caused any major performance degradations elsewhere.
Wen
Cosmin Oprescu 🇪🇺