Since some time, I mine twitter for good articles. I mine, I bookmark and never read them.
I am wondering if a kindle that is not connected to the internet can solve that.
IntelliJ is expensive for $200/year compared to other IDE among which most are free to use.
Yet, it has a great sway in the market. That really says something about its perceived productivity and market familiarity, isn't it?
Why Btree when you have binary tree?
good explanation here below: Essentially, pack as many children into a block because block based database engines read the entire block on access anyway. Btree with 4 levels and 500 children of 4KB can hold 256TB
https://t.co/ALWfVaxWrI
@deedydas I think you have a point. But I'll nuance it. In India, we're continuously trying to know where we stand in the status ladder compared to our neighbour, peer, cousin etc.
It's cultural. But it's not a way to gain or be denied entry into a select group.
Basic intro prompt engineering, its relations to consuming a GPT API.
GPT tends to not focus on the middle part of our prompt called "context" but at the end, we have to make it re-focus with something like "with that explanation, what I want is..."
https://t.co/q1ZzYCKMRD
Parallel streams use fork join model under the hood and are performant if the product of computations x no. of elements is high.
For simpler computation, fewer elements, parallel stream would have overhead costs making is slower than sequential stream.
https://t.co/sokULvAE0G
1 million blocks in MySQL, the engine would take ~ 20 block accesses on index key search. (binary search)
Ask ChatGPT: Is this because 2^20 =1million ?
ChatGPT: No
2^20 is indeed 1,048,576. ie. Log 2 x = N. N is total blocks.
ChatGPT: Yes, thanks for catching my mistake.
In certain scenarios, you may be faced with the question "Tell me about a challenge you had and how you dealt with it".
I think that the question is about what you *see* as challenges, more than how you *deal* with challenges.
A common misconception with LIMIT is that
select * from SomeTable LIMIT 10
will only return 10 records. Beware!!!
In a table with 10 million rows, this will do a full table scan and only then limit it to 10 records.
LIMIT is final "reduction" applied after all operations.
This is an actual job description merely asking for familiarity with LLM integrations not expertise.
Time to get your hands dirty with LLMs. No more excuses.
It is tempting during information gathering to open a hundred tabs.
It is important though to put all this into paper so that not only you but also others can chew on your research.
Gradually, you will see your notes turning into a story. And it will lead you to your next step.
Researching on a vendor, doing a slide deck on offerings and trade-offs, suggesting potential functional architecture and getting in touch with potential vendors to take a call on either going open source or picking something from a shelf.
All this is great skill.
If your interviewer asks you to do take-home assignment, don't shirk away from it if you have the time.
In one assignment, I learned about geo-spatial data type in MySQL. Querying this data from JPA was tough, so I had to fall back to JDBC.
There's a lot to learn in DIY.
@brunoborges it crashes once a in while and people complain that they don't receive notifications. But for most cases, it works and is heavily used in the industry.