eggie5

8 ms·

1.
by eggie5 1y ago
Adyen | Founding Research Engineer, AI | ONSITE (San Francisco) | Full-time Adyen is a global fintech platform powering payments, data, and financial services for companies like Meta, Uber, Spotify, and Nike. We’re profitable, publicly trad
2.
by eggie5 2y ago
Adyen | Sr DS & TL GenAI Team | Full-time | Madrid Looking for SR DS and Team Lead roles for our tech-hub in Madrid. We have A100 GPUs and lots of use-cases were ready to execute on. Read about some of our recent work in agentic flows:
3.
by eggie5 2y ago
What's _online_ RL for an LLM? Saw this on the llama 3.3 reports too...
4.
by eggie5 3y ago
ah, so observe the reward and then take a gradient step
5.
by eggie5 3y ago
if you have an NN that is probabilistic, how do you update the prior after sampling from the posterior?
6.
by eggie5 3y ago
just applied for GenAI position in AMS. Can you transfer visas?
7.
by eggie5 3y ago
to add state (memory), u can either: * inject the running chat log into the prompt * inject the summary of the chat into the prompt
8.
by eggie5 4y ago
LOL
9.
by eggie5 4y ago
I think the main criticism is portability of skills, ie everyone wishes ebay would just use react.
10.
by eggie5 4y ago
As Adams agreed, it was quite the perplexing philosophy. However, who won in the end? The firebrand Adams or stoic Jefferson???
11.
by eggie5 4y ago
everyone at ebay hates it -- the ebay blind is full of people hating on it
12.
by eggie5 4y ago
Years later, still puzzling over Thomas Jefferson's passivity at Philadelphia, John Adams would claim that "during the whole time I sat with him in Congress, I never heard him utter three sentence together" Jefferson, himself
13.
by eggie5 4y ago
yes and doesn't have traction inside ebay either :)
14.
by eggie5 4y ago
created at ebay and hated internally :)
15.
by eggie5 4y ago
in netherlands, for example, they (Goog) has to negotiate w/ the local-employee-backed Works Council. If they didn't bother setting one up, like Meta NL failed to, it will take about 3 months to do that. Then another 1-2 months fo
16.
by eggie5 4y ago
Interesting work on Online Learning: There are many empirical studies which show for feature hashing, a few collisions don't have a sig impact on perf ( https://youtu.be/ARjNMdCzN-Q?t=599 ). However, for some archs, the
17.
by eggie5 5y ago
how much do you get paid and what city?
18.
by eggie5 5y ago
levels.fyi to get some data points.
19.
by eggie5 5y ago
do interviews
20.
by eggie5 5y ago
just saw an offer in Berlin for 150 base and 320k RSUs coming out to €230k annualized TC!!
21.
by eggie5 6y ago
Grubhub (grubhub.com) | ML Engineer | Full-time Search Data Science team is looking for a senior-level ML Engineer to help us drive search at Grubhub. Search team builds personalized recommender systems and other data products drive search-
22.
by eggie5 6y ago
You missed a big part: they did a big NAS run to make it work.
23.
by eggie5 6y ago
Do what do attribute to the gains? Adaptive clipping? Or $$$ spent on NAS??
24.
by eggie5 6y ago
Grubhub (grubhub.com) | ML Engineer | Full-time Search Data Science team is looking for a senior-level ML Engineer to help us drive search at Grubhub. Search team builds personalized recommender systems and other data products drive search-
25.
by eggie5 6y ago
reading gulag archipelago so i go that reference
26.
by eggie5 6y ago
anyone still like classic notebooks better???
27.
by eggie5 6y ago
why?
28.
by eggie5 6y ago
I've gotten the question in ML eng interviews 3 times: implement sparse vector and the respective dot product op. I always go for the handy DOK method.
29.
by eggie5 6y ago
roam
30.
by eggie5 6y ago
Managing uncertainty with Distributions instead of point estimates

More ›