Pra Cha

ai, humans, coexistence, prosperity

[Pra Cha @PraCha98]

Read it again and again, a true compass for anyone

https://t.co/mfvn65kXQC

[Pra Cha @PraCha98]

@tamizh_tamizh Kind and Humble

[Pra Cha @PraCha98]

@Gosleepriya Skin😭

[Pra Cha @PraCha98]

My mom doesn’t know exactly what I’m studying; she just knows I’m doing an MS. While explaining Data Science and AI—how we make machines think and act like humans—I mentioned self-driving cars in SF and that I might commute to the office in one every day. She was surprised.

In a spontaneous conversation, I told her I might work on designing kids, ending cancer, forecasting the world, sending humans to space, growing lab meat, and more—whatever is possible with data and AI.

It felt like a manifestation, as manifestations are spontaneous, not deliberately crafted.

I did the same around 2020 when I said I’d go to or settle in Seattle, and my first MS admit was from UW.

I’m not here for short-term indulgence. I’m here for the long game.

[Pra Cha @PraCha98]

Give it a read

https://t.co/8o8zQfTD31

[Pra Cha @PraCha98]

X is the RL env for humans for hill climbing!

[Pra Cha @PraCha98]

@the_jasonsamuel Ideally it’s not just ML systems, anything data/analytics/ds works with fortune500 clients - whats now FDE (influenced by palantir)

[Pra Cha @PraCha98]

@the_jasonsamuel 2019

[Pra Cha @PraCha98]

@Kavya_learns @X https://t.co/NQZkZjSqwj

[Pra Cha @PraCha98]

@thatteluguqueer talk, talk, show the evidence, let them show the facts, keep it healthy discussion.

[Pra Cha @PraCha98]

Yes, that's how it should be done. Ask about expectations in terms of salary, learning, and experience (and vice versa). Agree if it's good, then proceed with a better interviewing method.

[Sumanth Raman @sumanthraman]

All those giving gyaan here that I should have asked his expectation before evaluation, is this the way it is done nowadays? So we first ask the resource what he expects and then if we can afford it we evaluate? Seriously?

[Pra Cha @PraCha98]

IMHO, the salary has to be based on experience, skills, and the role, not the mere percentage increment or 🍎 vs 🍊 comparison of companies.

When I left my first job at 14 LPA, the next role's max I negotiated was 32 LPA (with 3 YOE), and many were able to accept the number as I had experience and justification—so is everyone. Few offered max 25% on current CTC, and I didn’t even proceed with those interviews.

So, it's better to set and exchange expectations in the first HR round itself, to avoid wasting time and energy on further interviews and potential drama at the end.

[Sumanth Raman @sumanthraman]

Interviewed a candidate for a techie opening yesterday. CTC for the candidate with 4 years experience in current company is 7.2 LPA. Asked what the expectation was.
Candidate says 16 lakh. I said that's more than double current CTC. Candidate says yes that's what I want. Conversation ended soon after.
Increasingly feeling out of touch with today's generation.

[Pra Cha @PraCha98]

@fit_fr_nothing Oh okay, what about frugal but efficient ?

[Pra Cha @PraCha98]

Would there be a new employee review system trend that somehow evaluates frugality and proficient aspects of an employee against token consumption?

[Pra Cha @PraCha98]

long way to go!! i just used pg's how to do a great work blog to analyze my projects, plans, and various things through notion, github, my profile and what not ..

need to mint more useful outputs than "sharing thought process" as thoughts are latent and need more tuning for others to consume!

[Pra Cha @PraCha98]

@sidr29 Looks awesome ,

+ from Columbia too

[Pra Cha @PraCha98]

@debuglol Ngmi

[Pra Cha @PraCha98]

Based on various seminar courses I have attended (probabilistic models, causal inference, continual learning, advanced deep learning) which really helped me in terms of self-learning, I’m sharing a few notes.

Keep a map: where to start and where to end, along with a primer that introduces many new concepts. The map can start from ResNet and end with DeepSeek, but it should not list every paper or model, only the more seminal and “test of time” ones - no use of reading every other paper that pops up!

The primer can be a single book that you use to reference concepts. For example: Understanding Deep Learning by Simon Prince

For each paper, create: (a) a Cornell-style note or read report of less than a page, and (b) a Socratic dialogue / Q&A that “attacks” the paper from different roles: reviewer, scientific communicator, industry practitioner, or coder. I especially like the role-play of an archaeologist, whose job is to trace back the lineage of ideas and see how well they have withstood the test of time.

For example, take the idea of InceptionNet and ask: “Is anything from InceptionNet still being used in contemporary use cases or research?”

If you are interested, push it into an experiment or use the paper as a reference for one of your broader projects.

Keep crafting the overall arc of your learning. Over time, this compounds into a lot of depth.