Vihan Singh (@vihan13singh) 's Twitter Profile
Vihan Singh

@vihan13singh

funmaxxing

ID: 1472547556869632000

calendar_today19-12-2021 12:41:24

293 Tweet

1,1K Followers

3,3K Following

Vihan Singh (@vihan13singh) 's Twitter Profile Photo

The 10 principles to build an AI Company Capital: 1. This is the biggest opportunity since the internet. Bigger than cloud. Faster than mobile. 2. AI isn’t one market — it’s eating all markets. Software, services, labor, infrastructure. 3. The value is in the application

Vihan Singh (@vihan13singh) 's Twitter Profile Photo

There are 284 billionaires in India but over half of them come from one single state!! 🇮🇳 (yes, this is true, 1 single community contributes more to India’s GDP than most sectors combined) I’m talking about the Gujaratis. They are just 5% of the population, but contribute over

There are 284 billionaires in India but over half of them come from one single state!! 🇮🇳
(yes, this is true, 1 single community contributes more to India’s GDP than most sectors combined)

I’m talking about the Gujaratis. 
They are just 5% of the population, but contribute over
Vihan Singh (@vihan13singh) 's Twitter Profile Photo

oh, you’re into AI? then tell me about… Gradient Boosting, Random Forest, SVM, k-NN, Naive Bayes, KMeans, PCA, t-SNE, XGBoost, LightGBM, CatBoost, AutoML, Optuna, HyperOpt, TPOT, H2O.ai Sklearn, MLflow, Weights & Biases, DVC, ClearML, Comet, Neptune, Colab

Vihan Singh (@vihan13singh) 's Twitter Profile Photo

Still judging risk by the single worst drawdown? 📉 It’s only one outcome. 🔧 Back-tests can tweak it away. A 25 % max-DD won’t save you from a 50 % crash. Do this instead: ✅ Run Monte Carlo paths. ✅ Track volatility and tail stats (expected shortfall, ruin odds). Example:

Still judging risk by the single worst drawdown?

📉 It’s only one outcome.
🔧 Back-tests can tweak it away.

A 25 % max-DD won’t save you from a 50 % crash.

Do this instead:
✅ Run Monte Carlo paths.
✅ Track volatility and tail stats (expected shortfall, ruin odds).

Example:
Vihan Singh (@vihan13singh) 's Twitter Profile Photo

All my friends in SF (18 - 28) are working 80hr a week trynna retire their parents and all my friend in Delhi, Bengaluru, Mumbai, Jaipur, Kolkata (18 - 28) are partying 80hrs a week crashing their dads cars I dont think a big global co. is gonna come out of India tbh

Vihan Singh (@vihan13singh) 's Twitter Profile Photo

Everyone wants to be early. Early at a startup. Early to a trend. Early to AI. But no one talks about what being early actually looks like: • No instructions • No social proof • No market map • Just instinct, curiosity, and a ton of failure Early doesn’t feel cool. It

Vihan Singh (@vihan13singh) 's Twitter Profile Photo

Hiring rn feels so much like a lottery draw for both the company and the candidates. There has to be a better way to cut through the BS and get the signal from the noise.

Vihan Singh (@vihan13singh) 's Twitter Profile Photo

🎥 HIRING: VIDEOGRAPHER (big plus if also EDITOR) 📍 Gurgaon 💼 Full-Time 💰 ₹4–8 LPA (video only) // 8-14 LPA (vid+edit) 🔧 Responsibilities * Shoot and edit high-impact content: reels, shorts, founder docs, gym/trading B-roll * Own end-to-end production—planning, filming,

Vihan Singh (@vihan13singh) 's Twitter Profile Photo

🎥 HIRING: Media & Content Team ROLES: 1. Content/Script Writer 2. Videographer 3. Post Production (Editor, Graphics, Motion Graphics) 📍 Gurgaon 💼 Full-Time (In-Office Only) 💰 ₹6-12LPA 🔧 Responsibilities: create impactful scroll stopping content ✅ Requirements: proof

Vihan Singh (@vihan13singh) 's Twitter Profile Photo

gradient descent is a great decision making framework with AGI coming. the key principles is as such. stochastic gradient descent guarantees results even under uncertain conditions. it does this by having each step build upon the other in roughly the direction of the bottom

Vihan Singh (@vihan13singh) 's Twitter Profile Photo

Want to do CS? Learn maths ML? Learn maths VLSI? Learn maths Theoretical physics? Learn maths Quantum computing? Learn maths Cryptography? Learn Maths Robotics? Learn Maths Quant? Learn Maths Finance? Learn Maths

Vihan Singh (@vihan13singh) 's Twitter Profile Photo

i don’t think you understand you train a small model on scikit-learn it gets 80% accuracy you fine-tune a huggingface transformer it gets 90% accuracy you build a custom data pipeline it gets 93% accuracy you ensemble three models with voting it gets 95% accuracy you throw in