Joined March 2018
- Tweets 109
- Following 485
- Followers 30
- Likes 885
7 Photos and videos
Registration is now open for the 5-Day AI Agents: Intensive Vibecoding Course with @Google π
This no-cost course is designed to help builders learn how to design, build, and use AI agents using the latest concepts, technologies and skills.
ALT Promotional graphic for a "5-Day AI Agents Intensive Vibecoding Course with Google" by Kaggle and Google, scheduled for June 15 - 19, 2026. The image features playful illustrations of a laptop, coding icons, and abstract geometric shapes.
12
131
693
67,906
Rajend retweeted
Mar 25
5 days left to apply for the AI Engineering Buildcamp scholarship.
I offer several full- and partial-scholarship spots.
If you're motivated to learn but cost is a barrier, apply here: forms.gle/qAex9yTkZqypJMk67
Application deadline: March 30, 2026
If you know someone who'd benefit, please share and tag them.
ALT Scholarship application details for an 8-week AI Engineering Buildcamp focusing on hands-on learning and practical AI projects.
8
41
234
21,485
12 Dec 2025
Iβm on pg 252 of 'The Accidental CTO' - the content is brilliant with great analogyπ©·
but the structure in the second half becomes confusing and messy, makes it difficult to read and breaking my flow. Here are all the issues I found in the current PDF π
@subhashchy
2
49
12 Dec 2025
Despite the structural issues, the content is fantastic. A cleaned-up PDF with proper chapter order, numbering, and missing sections added would make the reading experience as smooth and intuitive as the concepts you explain. Thank you for writing this book. @subhashchy
45
12 Dec 2025
2. After the Docker chapter, the entire build-up suggests Kubernetes should come next but instead, 2 unrelated chapters appear before Kubernetes.
3. Chapters appear out of order:
chapter 13
chapter 16 - 238
chapter 15 - 245
all are skipped
chapter 19 - 269
chapter 16 -280
1
28
28 Nov 2025
Day 17/100 β #100DaysOfML π
- Bias : model too simple, misses patterns (underfitting)
- Variance : model reacts too much to small data changes (overfitting)
- BUVO: Bias Underfitting, Variance Overfitting
β
Done L1 & L2 reg @codebasicshub exercise.
#MachineLearning #AI
19
25 Nov 2025
Day 16/100 - #100DaysOfML π
- Learnt how L1 (Lasso) & L2 (Ridge) regularization help reduce overfitting by penalizing large coefficients.
- Also revised causes & fixes for overfitting/underfitting.
Practiced Labs for it.
#MachineLearning #AI
2
29
25 Nov 2025
Underfitting causes:
1. model too simple
2. bad feature engineering
3. not trained enough(less epochs)
4. excessive regularization
Fix: better features, more training, more complex model
13
25 Nov 2025
Overfitting causes:
1. too many features
2. poor model choice
3. little data
4. no validation
5. no regularization
Fix: better features, more data, k-fold, apply regularization
8
19 Nov 2025
Missed posting for a few days, but Iβm back on track!
This week I learnt:
- Linear Regression
- Multiple Linear Regression
- Polynomial Regression
Today I completed the exercise & lab for Poly Reg.
thanks to #campusx Linear Regression playlist.
#MachineLearning #AI #codebasics
ALT home made linear regression class :)
1
2
45
5 Nov 2025
Day 14/100 β #100DaysOfML π
Today I learnt:
Practiced 1-0 (One-Hot) Encoding for nominal data.
Reduced multicollinearity by removing one dummy column.
Trained & evaluated the model after encoding.
@codebasicshub
#MachineLearning #AI
28
3 Nov 2025
Day 13/100 β #100DaysOfML π
Today I learnt:
Applied MSE, MAE, and R2 Score to evaluate model performance.
Multicollinearity : when features are highly correlated.
Dummy Variable Trap in 1-0 Encoding can cause it, fix: remove 1 column.
@codebasicshub
#MachineLearning #AI
35
2 Nov 2025
Day 12/100 β #100DaysOfML π
Learnt about R2 score and done lil lab
Ps: gonna resume my 100 day journey from today. Also i passed Oracle Foundation Ai cert examπ
#MachineLearning #AI #oraclecertificationprogram
44
15 Oct 2025
Iβm taking a break from ML today to recover from a cold and come back refreshed tomorrow.
#100DaysOfMl #ai
42
14 Oct 2025
Day 11/100 β #100DaysOfML π
Learnt why MSE > MAE for GD
MSE (xΒ²) :
- Best when few outliers.
- smooth, differentiable (fβ(x)=2x).
- GD finds minima easily.
MAE (|x|) :
- better with many outliers.
- not smooth, undefined at 0.
- harder to optimize.
#MachineLearning #AI
51
13 Oct 2025
Day 10/100 β #100DaysOfML π
Learnt:
Gradient Descent : finds the best-fit line by adjusting slope & intercept to reach the global minim
Manually found the best-fit line using MSE and partial derivatives.
min max scaling: bring features into 0β1 range.
#MachineLearning #AI
28
4 Oct 2025
My system got corrupted so evrything is delayed(currently fixing) also exam going on.
Hopefully will post my progress on ml from tommorow or soon after exam.
23