dongobread

6 ms·

31.
by dongobread 3y ago
I tried this and collective wealth does increase: https://imgur.com/a/Zc2aApj The median person does get poorer over time though, a small percent get extrmeely wealthy: https://imgur.com/a/v2AVyrv
32.
by dongobread 3y ago
I have used much more Python than R in my career and I still find I'm able to do most analytics work about 2x faster in dplyr/tidyverse + data.table than in the pandas ecosystem. Polars is better but still not comparable to dplyr.
33.
by dongobread 4y ago
All these companies selling CRUD wrappers around ChatGPT + an embeddings database seem like awfully cynical cash grabs to me. Would love to see YC sponsor some companies trying something more novel on the LLM side.
34.
by dongobread 4y ago
They'll release training logs but no model file or data? How does anyone trust that this isn't just overfitted garbage trained directly on their benchmarks?
35.
by dongobread 4y ago
If they really want to get the jump on Microsoft, they should open source the model parameters and the weights, similar to how Stable Diffusion grabbed all the interest from Dall-E.
36.
by dongobread 4y ago
I would agree that these tools are almost inevitably going to fail. If anything, they'll only improve the models' abilities to handle edge cases accurately. However, I'm surprised at how little empathy is being displayed towa
37.
by dongobread 4y ago
I worked in a research capacity in the voice assistant org of a big tech company until very recently. There was a lot of panic when ChatGPT came out, as it became clear that the vast bulk of the org's modeling work and research essenti
38.
by dongobread 4y ago
It does not surprise me at all that performance ratings are correlated with office presence, but not for the reasons they imply. The performance evaluation system at Meta is heavily biased towards political visibility and grandstanding and
39.
by dongobread 4y ago
There have been a least half a dozen companies and countless FOSS projects attempting to do the exact same thing. What makes this special?
40.
by dongobread 4y ago
All the more reason to move on to Hetzner or Linode.
41.
by dongobread 4y ago
For me, the ease of migration is huge as well as the ability to run WP on extremely low-end VPS with limited RAM.
42.
by dongobread 4y ago
It's very unclear to me why the writer ran everything through GCP and Colab. Having everything on a single environment with near-100% uptime is certainly still possible by renting a VPS, which is usually a much cheaper alternative anyw