temikus

I’ve personally found MiMo models a hit and miss. I have some personal agentic projects and I found them to hallucinate hard at least 10% of the time. And do so in pretty sinister ways - making up people, names, places, etc. I switched back to Kimi for now.

I’m in senior management and my take is - this is one of the very few times in our industry when the companies are actually both encouraging and investing in L&D at an unprecedented rate. When do you last remember a company being comfortable throwing 10,000$+ a year for you to learn a new skill and experiment at your own pace? This is why I’m encouraging my engineers so much - it’s free training! Take it while it lasts!

That was my thinking as well, in Australia this would be a b-double tandem, requiring extra endorsement (multi-combination license).

This is very very likely. Source: worked on the storage backing Google Drive. There’s layers upon layers of backups.

I feel Evolve skateboards are next. Remote drop-outs and start/stop issues have been known in community for ages. I miss Boosted - they really cared about rider safety :(

This is a very good comment. I think the article presumes lower level roles. Very senior comp packages are wildly different so you always go into the convo with “I expect at least X, can you meet my number?” Otherwise you’re that person that tries to negotiate a 250k position up to 500k which is not going to happen.

I view this not as a “you’re making me wait” thing but rather being kind to those of us who have anxiety. “Hi.” without any follow up can really derail a struggling person.

Statistically speaking 2/10 have an ongoing mental struggle so I try not to make the “hi” transactional and immediately follow up with the point in the same message.

Ok, so I actually had to deal with this.

Pick some sort of standard, for example CAIQ and have an always-up-to-date version of it. You’d be surprised how many customers would accept it if you tell them “hey - we use a standard - is this acceptable?”

After that - figure out what certifications will be advantageous. Then automate, automate, automate with something like Hyperproof/Vanta. You will still need a compliance person or more likely a team at that point, so those certs have to unlock some serious money. Otherwise - just stay on top of VSA’s until running a compliance program makes sense.

Just don’t fall for the baseless “SOC2 equals enterprise customers” spiel. Analyse your pipeline and regulatory environment and make a call based on that. So many startups spend millions running a compliance program that brings in thousands.

Not the OP but I am sensitive and sadly found the MiniLED to be unusable as well. Had to switch from a 15’ MBP to the low-cost 13” with the old-school 500nit display.

Same. There are now devices on the market that help find out sensitivity risk, I use this one when I’m going for hardware shopping: https://flickeralliance.org/collections/tools/products/opple...

Depends on whether they threw out the key and crypto shredded. Contact support immediately and keep escalating like your life depends on it. You _might_ be able to recover it but chances are slim.

I worked with some. General things I picked up from them and try to emulate now that I’ve become one:

- Clearly state your values and stick to them - it’s much more comfortable to people to work with a predictable leader and builds trust over time.

- Provide rich context to the team and ask for business outcomes, not specifics.

- Communicate, communicate, communicate - especially if you work in a remote team/company. You need to have a constant dialog going with the team - it’s up to you how, however - there’s a lot of different ways to facilitate.

- Remember your promises and provide people with resolutions to issues they raise, follow up! I know it sounds self explanatory but so many managers just plain forget the things you ask immediately after your 1:1’s.

- Do your 1:1’s for gods sake! Weekly preferably.

I do agree with the general premise but there are a lot of passages that kinda trivialise a lot of complicated phenomena, e.g.:

> Much like victims of childhood hardship or abuse can find escape in fantasy books, victims of enterprise programming or freelance web development can find their escape in solving imaginary problems.

It’s not a bad article. Only feedback I would have is about this passage:

“”” If you routinely pass in a dataset to AI, it can identify anomalies and watch to see if it settles or even reason why the anomaly occurred in the first place. “””

This is not a great use for LLM’s. AI? Sure. But not this type of it.

Anecdotally - Reddit is not really useable for me today. More than half of the communities I frequent are offline so the home feed is just the odd ones that are.

Because killing off the ecosystem has gone so well for all the other social networks. They’re going to really run the company into the ground :/

Google Translate / Waymo is fair, I totally forgot about those ones.

Gmail Autocomplete, Photos Classification, YouTube recommendations are (relatively) simple models and don’t differentiate from the market much.

The thing that bothers me is that Google has been promising market-beating ML-as-a-service for ages which never materialised despite billions being poured into it.

Google Prediction API has been a great start in the area. Then Google stopped improving it and killed it with very little notice and without providing an alternative.

It’s not about ChatGPT replacing Google IMO but the fact that decades of very expensive AI research at Google produced relatively little apart from impressive looking papers. If I was a major shareholder I would be asking some uncomfortable questions to Sundar right about now.

EDIT: Yes, there is TensorFlow but that is not really a product of the core AI teams, rather than a need for better tooling from them.

TIP: this is why you set up Google Alerts[1] on your name and any common nicknames/usernames you use.

TIP: this is why you set up Google Alerts on your name and any nicknames/usernames you use often.

Hm, I wonder if “don’t eat Hare’s head” advice was sprung because of Tularemia which can transmit to humans.

I’m still surprised that it’s a revelation to many managers.

“What do you mean an oppressive environment is not conducive to good team dynamics?”

Oh man, this is major. I would’ve loved to have something like that 10 years ago when CPU was a bit more precious. Still very useful today, just not to the same extent.

Not the worst decision considering how VIX is looking.

I’m not saying bad things about the author rather that the authors don’t hold proper credentials so the reader should examine it carefully.

How is this ad hominem? This is like an article about cardiac health that says that we need to immediately stop administering Aspirin to the entire population written by an OBGYN with a plumber that sells Tylenol as a co-author.

This is essentially wiring people’s jaws shut. Not sure how effective this may be outside of research.

“Tell me you’re rich without telling me you’re rich.”

Yep. Not a single immunologist in authors. The last co-author is Chief Medical Advisor, Truth for Health Foundation, which is a religious antivax nonprofit, I quote directly from their about page:

“We strive to honor God and Jesus Christ in all we do as we conduct business, and to follow Biblical principles exemplified in Jesus’ Healing Ministry throughout the New Testament as we implement health programs for the people we serve.”

I’m in Australia and people are starting to get back into it. Cannot say about Bay Area as it’s been a while since I’ve visited.