Kapil Zadpe

All I do is like and retweet |
Freelancer | Sr. MLE @EliLillyandCo | IITKgp'23

[Yun-Ta Tsai @yunta_tsai]

Many people think any given ML project is 99% training.

In reality, it’s 50% evaluation, 40% data cleaning, 8% integration, and 2% training.

The first two set the noise floor for learning. No ML magic matters; the model cannot lower the noise floor, as that’s the optimal bound of Shannon encoding of your data.

Thus, not a single day goes by without me thinking about ontology. Even the old labels have to be constantly reviewed.

[Malvin @ManOfFocus_]

Once you realize that anything can happen; sickness, death, lose your job... Literally, anything in the blink of an eye, you become very humble. Tables turn and that's how crazy life can get.

Always stay humble, and be grateful.

[Mercy @Frankhavemercy]

One of the most overlooked forms of wealth is having complete ownership of your time.

Waking up and knowing nothing about your day will be decided by someone else is priceless.

[blue @bluewmist]

Uncommon advice: If you don't know what to pursue in life right now. Pursue yourself. Pursue becoming the healthiest, happiest, most healed, most present, most confident version of yourself. Then the right path will reveal itself.

[Tai Lopez @tailopez]

Marcus Aurelius was right. You will lose friends you will lose lovers you will lose comfort but if in losing them you find yourself you have gained more than kings.

[Ganesh Vanare @haram_khor_]

Photos I Took at Raigad Fort.

[Tech with Mak @techNmak]

These engineering blogs have leveled up my tech skills more than any bootcamp, course, or conference.

Here are the ones worth bookmarking:

[Prompter @PromptLLM]

you need to start visualising

[Āsurā @iAsura_]

Indian thugs sell you mirages like blissfulness, shanti, love etcetera, whereas the world is a hunting ground full of predators.

Anything that hides your predatory self is a bait to convert you into a prey.

Don't be someone's food. Eat.

[प्रशांत @Im_prashant812]

A Solo Guide to Annapurna Basecamp(4130m)🇳🇵

Distance ~ 60kms
Duration - 6 days
Budget ~12k INR (Pokhara to Pokhara)

Day wise details 🧵

[THE SKIN DOCTOR @theskindoctor13]

Why are we living? What is the purpose? What comes after this life? Why does anything exist at all? There are no universally accepted scientific answers to these questions.

The most common material explanation can feel emotionally unsettling: that life is simply a temporary biological process. One day it ends, consciousness disappears, and you cease to exist. Not only do you die, but your awareness itself vanishes. You won't even know that you no longer exist because the very thing that experiences, remembers, and feels would be gone.

For many people, that thought creates an existential vacuum. Belief in God fills that vacuum with meaning. Life becomes a journey rather than a random event. Suffering can have purpose. Death becomes a transition rather than absolute erasure. Instead of "I exist briefly and then disappear forever," it becomes "I am here for a reason and my existence has meaning beyond the physical."

Whether God exists or not is a separate debate, nor am I urging anyone to believe in God, but psychologically it is understandable why faith can make people happier. It offers answers where uncertainty can feel frightening, and certainty itself gives the mind rest.

[Sadhavi Khosla @sadhavi]

The Ashtavakra Gita is the only text that is completely ruthless about the role of effort in liberation. Every other text prescribes practice. Patanjali prescribes Abhyasa & Vairagya. The Bhagavad Gita prescribes action without attachment. Even the most radical Advaita texts suggest some form of inquiry. Ashtavakra stands completely alone. He says effort toward liberation is like running in dark. The more sincerely you run the further it recedes. Not because liberation is impossible but because it is not in the direction of effort. It is in the direction of complete and total relaxation into what you already are. This is the most liberating teaching ever given. Because it removes every strategy, every technique, every spiritual achievement. And leaves only the bare fact of what is already here.

[‎Wojak Codes @wojakcodes]

you can avoid responsibility all you want, but you can’t avoid the consequences of avoiding responsibility.

[Arpit Bhayani @arpit_bhayani]

pro tip: get good at sounding confident even when you know nothing. ask questions when essential, and trust yourself to figure the rest out.

because everything is figureoutable.

[Vikram Pai @vikpai]

This is it.

Everything I’ve learned working with India’s top D2C brands for 2 years

From: @ReferRush
To: Every Indian D2C founder.

0. Distribution matters more than almost everything.
1. Your product can’t save weak distribution.
2. WhatsApp is the most important channel in India.
3. Referrals work extremely well in India when pushed correctly.
4. Most brands barely activate their referral programs.
5. Your existing customers are your cheapest acquisition channel.
6. Consumers trust people more than ads.
7. Every extra click destroys conversion.
8. The simpler the flow, the more revenue you make.
9. Most brands send too few nudges.
10. Consistency beats campaign spikes.
11. Good offers outperform pretty creatives.
12. Founder-led content works absurdly well right now.
13. Short-form video is still underpriced attention.
14. Most brands underestimate repeat purchases.
15. Retention is where real profit is made.
16. Email still prints money.
17. SMS still works.
18. WhatsApp open rates are unfair.
19. Fast replies increase conversion dramatically.
20. Cash flow matters more than vanity growth.
21. Most D2C problems are operational, not marketing.
22. Shipping experience affects retention more than founders realize.
23. COD creates both growth and pain.
24. The best brands obsess over post-purchase experience.
25. UGC beats polished ads most of the time.
26. Real customers are your best creatives.
27. Most consumers need multiple reminders before buying.
28. Reviews matter more than founders think.
29. Communities compound brand growth.
30. Great copy matters more than founders think.
31. Most landing pages are too complicated.
32. Mobile-first is mandatory in India.
33. Speed matters. Slow websites kill conversion.
34. Your first 100 loyal customers matter more than your first viral reel.
35. Strong brands repeat the same message over and over.
36. Most brands try too many things at once.
37. The best operators focus on a few channels deeply.
38. Influencer seeding works if done consistently.
39. Affiliates need constant nudging and relationships.
40. Brands that win make sharing easy.
41. The next generation of Indian D2C brands will be built on community, retention, and distribution.

Start there.
India’s D2C wave is still early.

[Karan🧋 @kmeanskaran]

I have been freelancing and consulting in AI/ML since 2023 alongside full-time jobs, and I hardly know a month without gigs.

I am transitioning from freelancer to consultant.

I have never used Upwork or Fiverr because they are saturated, overcrowded, and rarely offer actual ML work and many sell the same services for less.

I got every freelance opportunity through LinkedIn and some content creation through X. This experience is different and thrilling; everyone should give it a shot.

Very few people get this exposure because of an optimized LinkedIn profile and strong proof of work with good documentation.

Start posting your projects and document them cleanly. Documentation should be 40% technical and 60% focused on the problem statement, challenges faced, and how you tackled them.

Add your freelance experience to your LinkedIn About section. If you have none, use personal projects as proof of work, but be ready to communicate and solve issues in interviews.

Freelance work can include content creation, coding products, designing architecture, teaching, consulting researchers, etc.

What matters is trust and quality documentation of projects.

Transitioning to consulting takes time. First stabilize small gigs and build a freelance portfolio. Consultation requires prior experience, business-level work, high trust, and strong pace.

Don't chase money in freelancing. Win clients' hearts through quality work, communication, and trust. You will eventually attract high-paying clients.

For consulting, you need no more than 2-3 clients max, as it is mostly decision-making and requires sharp focus beyond coding.

I never work with competitors of my current organization and avoid jobs requiring coding. I focus on distribution work, content writing, SME work, and decision-oriented tasks like designing systems.

I never take on work similar to my job. I am exploring SEO, Google Business setup, and ML.

The process is complex to explain, but I tried my best to cover it.

Comment your thoughts. I'd love to discuss.

[Sergey Nazarov @sergeynazarovx]

We used to go to a special website, ask strangers for help with programming, and get humiliated in return

[Vatsal Sanghvi @vatsal_sanghvi]

do everything that you can to increase the surface area of luck

a job, a startup, a relationship

whatever it is - do everything, shoot the wildest of shots, be shameless and do things with intent

[priyanshu.sol @priyanshudotsol]

btw this is the link -> https://t.co/88cKUkfqtV

[𝗦𝗮𝗻𝗷𝗲𝗲𝘁 𝗦𝗶𝗻𝗴𝗵 𝗕𝗵𝗮𝗱𝗼𝗿𝗶𝘆𝗮 @showupsanjeet]

You don’t realize how much progress you’ve made

until you look back

and see

the version of you
who once struggled with things
that now feel easy