odyssey7

13 ms·

241.
by odyssey7 1y ago
If they've managed to escape Python for machine learning tasks then it's worth it imo.
242.
by odyssey7 1y ago
I've experienced this while leaving a different company that used a rare language. It's a tough situation being experienced in <peculiar language for Company A> when you need to ace technical interviews in <mainstream la
243.
by odyssey7 1y ago
I don’t buy the economic argument favoring memory-unsafe languages. There are fast memory-safe options. Legacy codebases can eventually become more expensive to maintain than to rewrite. What is the economic cost of an Achilles’ heel when c
244.
by odyssey7 1y ago
I don’t know the details as specifically, since I haven’t been able to justify investing my efforts in the non-flagship ecosystem within the TensorFlow project after it previously added its Swift version to the Google Graveyard, but TensorF
245.
by odyssey7 1y ago
There are choices at multiple levels. Yes, today’s ML engineer has practically no choice but to use Python, in a variety of settings, if they want to be able to work with others, access the labor market without it being an uphill battle, an
246.
by odyssey7 1y ago
“Specious” meaning what exactly?
247.
by odyssey7 1y ago
The same reason it’s good for web servers. It excels at even-driven applications.
248.
by odyssey7 1y ago
AI engineers will literally invent a new universe before they touch JavaScript. The death knell for variety in AI languages was when Google rug-pulled TensorFlow for Swift.
249.
by odyssey7 1y ago
There’s a practically infinite space of possible innovations. The fact that any part of it has ever been explored is just down to luck—and that’s ignoring the fact that once evolution discovers an innovation, survival of the fittest is stil
250.
by odyssey7 1y ago
You’re comparing the intelligent design of your job search to evolution.
251.
by odyssey7 1y ago
That cost and benefit trade-off emerged as an opportunity only through luck and it survived only through luck, even if the odds were in its favor.
252.
by odyssey7 1y ago
That claim jumped out to me as well. Evolution is supply and demand, cost and benefit, capacity and constraints, none of it balanced by anything apart from luck.
253.
by odyssey7 1y ago
Most doctors have insurance for covering their mistakes. We might expect an AI medical startup to pay analogous premiums when it’s paid analogous fees.
254.
by odyssey7 1y ago
I don't know, I once messed up a Big Tech interview question that was about enumerative combinatorics.
255.
by odyssey7 1y ago
Python in general and its ecosystem are some of the biggest roadblocks to building real-world AI systems, in my opinion. Python is a scripting language that makes it easy to produce student-quality work. Academia adopted it for its pedagogi
256.
by odyssey7 1y ago
I would say they don’t need to be perfect, but mere stability seems to be a thing of the past. The millennial generation has seen repeated waves of economic crises and layoffs and the invention of a gig economy. Mere stability is lacking. A
257.
by odyssey7 1y ago
The war was likely to eventually end. The price of food will continue to increase exponentially, at some variable rate of inflation, forever. Skilled professions were likely to remain useful after the war. AI makes their future value seem u
258.
by odyssey7 1y ago
They're more like engineering than pure math because pure math hasn't solved the n-body problem.
259.
by odyssey7 1y ago
There’s no way that we happen to exist in just the first 10^78 years of eternity.
260.
by odyssey7 1y ago
I don’t buy it. I suspect that a reason Google Cloud remains a laggard in market share is that Google’s core business, as an advertiser that feeds off user data, gives parts of the available market the heebie jeebies. The monopoly is possib
261.
by odyssey7 1y ago
Maybe they didn’t invent the problem, but the admissions priorities are theirs to repair
262.
by odyssey7 1y ago
And yet CS grad programs seem to care about only the math section of the GRE
263.
by odyssey7 1y ago
Never waste a good lemon
264.
by odyssey7 1y ago
Okay this is a fair rebuttal.
265.
by odyssey7 1y ago
You'll be disappointed to learn that most new medication approvals are obtained by pharmaceutical companies, rather than academia.
266.
by odyssey7 1y ago
A leader with absolute authority over the company isn’t necessarily a good thing. Publicly traded companies have accountability to more stakeholders in additional ways. A private company isn’t necessarily better for being able to avoid that
267.
by odyssey7 1y ago
If professional science has gotten to be more about trust than reproducibility, then professional science cannot be trusted.
268.
by odyssey7 1y ago
Ever heard of an engineering firm?
269.
by odyssey7 1y ago
Good news, or just deserts, the feeling is mutual. A large part of the electorate distrusts academics and is in turn dismissive of academics’ priorities and recommendations. It’s correct for members of the public to distrust institutions th
270.
by odyssey7 1y ago
If you notice a company lie while recruiting you, when should you expect them to start being truthful or fair? What other lies might await the candidate?

More ›