dd-sharma

I forgot to add the reason behind developing this methodology. Over the years in my career, I have seen too many pay to play software product evaluations and quadrant placements. We wanted to create something open, transparent, and easy to understand.

There's no failure if you keep doing what you really believe in. I feel labeling an outcome as a binary success or failure is not fair. For my first tech startup, I had lofty goals - a large scale company within a few years. It didn't get to that point and after a few years, it felt like it was becoming a lifestyle company. So, from the perspective of that specific goal, I failed and I felt that way at that time. But I feel I set that goal without considering what will happen on the journey, what all factors will be in my control and what won't. Even if I considered them, I wouldn't have known numerous factors. I faced some unforeseen challenges - some real big ones e.g. the great recession and investor capital drying up overnight. It was a real grind for a few years. But looking at that journey, I acquired at least 3x learning/experience what I would have learnt in my corporate job for the same duration. I acquired confidence in my ability to create money. And I feel it also made me a more humble person. So, my suggestion is if you really believe in what you're doing, don't give up. Besides the end goal, try to enjoy the journey too. Good luck!

I maybe in minority on this. I can focus only on one thing at a time. I fail miserably if I try even background music. It would play but I probably wouldn't register. And if I hear the music then my work will take at least twice as longer.

I guess people working at model companies. There is a lot of investor money floating around in different forms. Familiar with circular financing? Here's what seems to be happening. Chip company or a cloud provider invests equity/capital into an AI lab => AI Lab uses that funding to place massive GPU orders back with the chip maker or using compute of the cloud provide => Chip maker/cloud provider books revenue. Higher the revenue, higher the stock price :-).

I don't think the role of an engineering leader has changed much. I view AI in different forms as a tool to help us do our job more efficiently and perhaps more effective in doing the grunt work. You still have to do all the things you were doing before - planning, budgeting, ensuring quality and on time execution while keeping the team(s) motivated. One Bay Area peer group of Engineering Leaders that I was involved in is SFELC (San Francisco Engineering Leadership Council?). That time there used to be after work gatherings for leaders to network and discuss issues. I am not sure what's it current status.

I have been involved with my local high school for past 3 years. Like everybody else school districts are struggling to figure out the role of AI. Some of the teachers I know use AI tools to detect if students used AI for their homework assignments. When an assignment is flagged, this leads to a challenging exchange between the teacher and the student. Having said that, I would say school systems don't yet have a plan. There are community, legal and regulatory implications too.

This may not be the answer you want to hear. I put all my attention on enjoying every bite. Think of mindful eating. I find it stress buster. I don't recall when I did overeating.

Our initial need is to evaluate text generation only. Later we will be evaluating the detection of other types of content too (images, videos, and audio). Any input on them will also be very helpful. Thanks in advance, again.

Really? They raised a whopping $450 million in series A earlier this year. I doubt these investors would be comfortable waiting that long.

About 4-5 years back this was an experimental discussion here but looks like one company (Inertia) is partnering with Lawrence Livermore Lab to commercialize it now. It is considered to be cleaner and safer than nuclear fission based energy. While it neither uses radioactive material as input or output, it is not completely risk free. For example, the reactor wall over time becomes radioactive, although with a much shorter half-life. But certainly not at the level of Chernobyl. This could be a boon for EVs and AI data centers. But according to Inertia, the full-fledged operation is still 5-10 years away. I wonder how the energy landscape will shift as it becomes mainstream.

No, I don’t think model owners want to steal my data or IP. They already have too many smart people. The point is how it going through their training pipeline could potentially get exposed to actors who use their models.

For serious work, depending on the complexity of underlying functionality, I would choose the appropriate architecture and technology. For example, for a complex and large scale system I would go ahead with a 3-tier architecture; Frontend - React/Typescript, Java/Python/GO based Services for middle tier, and Relational or NOSQL database depending on the data throughput and integrity requirements. For simpler websites there are many different options based on how much time you want to invest.

I love using WP for my blog and I've a self-hosted version. In your question "new websites for beginner" indicates that the user is a beginner and wants to build websites. If websites have simple and static content that don't involve any serious stuff (e.g. e-commerce) then WP is probably ok. But for serious work i won't use it.

Sorry, I guess I missed the point. Looks like I got carried away by my experience.

Last week Pew Research recently reported an interesting paradox. Nearly half of U.S. adults say they use AI tools such as chatbots at work or at home. Yet respondents were significantly more likely to expect AI's long-term impact on society to be negative than positive. Interestingly, the younger folks (18-29 years old) were using AI more, yet 48% of them have a negative long term view. I know many technologies become more trusted as people gain experience with them. AI may be following a different path. I'm curious whether others think this is primarily:

- A transparency problem - A reliability problem - A privacy problem - A media perception problem - or Something else entirely

Several years back I lived in Tokyo for a couple of years. I like learning the local language. I could mostly get by with the spoken Japanese (Nihongo). Tried to learn the script but failed miserably. It's a combination of Hirgana alphabets, Katakana alphabets and to top it all 10s of thousands of Chinese Kanji. Loved my stay, very polite people. I think their vocabulary doesn't have the word impolite (polite and less polite). I would say Japanese script is one of the hardest if not hardest.