Weโve been building. Now itโs time to show it.
TECHRISE 3.0 DEMO DAY is here ๐ฅ
Sat 15th August, 2026
Hosted by @LearnFactoryNG + @KohiDojo
Expect: Live project demos, networking, and the next wave of African tech talent.
See you on the Road To Demo Day ๐
#TechRise3#DemoDay
New experiment: Does transforming the TARGET variable improve house price prediction?
Tested 6 versions on California Housing:
Original vs Log vs Sqrt vs Box-Cox vs Yeo-Johnson vs Quantile
Code: [https://t.co/to55iYQ4Jm]
#DataScience#ML
New project drop ๐
Built ML models to predict stroke risk from clinical data.
Compared: Logistic Regression, NB, KNN, SVM
https://t.co/gX9b99BRYT
#100DaysOfCode#DataScience#MachineLearning
New Project: KNN Classification โ
Baseline k=3: 96.6%
After tuning k with CV: k=1, 100% accuracy ๐
Lessons learned:
1. Scale your data
2. Tune your hyperparameters
3. Evaluate properly
Code: [https://t.co/df4xQgXvAc]
#Python#MachineLearning#DataScience
Built a Python fruit guessing game today ๐ฎ
Computer picks: grape, mango, apple, tangerine, pineapple
You guess โ Wrong? Try again โ Right? You win
Not Kaggle. Not a model yet. Just me learning loops + random.choice()
Small reps > no reps. Code on GitHub ๐
#Python#Build
Built a Python fruit guessing game today ๐ฎ
Computer picks: grape, mango, apple, tangerine, pineapple
You guess โ Wrong? Try again โ Right? You win
Not Kaggle. Not a model yet. Just me learning loops + random.choice()
Small reps > no reps. Code on GitHub ๐
#Python#Build
Built a Python fruit guessing game today ๐ฎ
Computer picks: grape, mango, apple, tangerine, pineapple
You guess โ Wrong? Try again โ Right? You win
Not Kaggle. Not a model yet. Just me learning loops + random.choice()
Small reps > no reps. Code on GitHub ๐
#Python#Build
Iโm a Mechanical Engineering grad diving into AI/ML, and Iโm documenting the whole journey publicly.
Not because I have it figured out. Because I didnโt, and I know how overwhelming the jump from ME to AI feels.
If youโre an investor building in Industry 4.0, IoT, or smart manufacturing: letโs talk. Iโm learning fast and building for real problems, not Kaggle competitions.
Iโm a Mechanical Engineering grad diving into AI/ML, and Iโm documenting the whole journey publicly.
Not because I have it figured out. Because I didnโt, and I know how overwhelming the jump from ME to AI feels.
My goal for the next 90 days:
Ship 3 end-to-end ML projects that solve actual mechanical problems. Document the wins, the bugs, and the exact commands I used. Everything open-source.
Iโm a Mechanical Engineering grad diving into AI/ML, and Iโm documenting the whole journey publicly.
Not because I have it figured out. Because I didnโt, and I know how overwhelming the jump from ME to AI feels.
Iโm a Mechanical Engineering grad diving into AI/ML, and Iโm documenting the whole journey publicly.
Not because I have it figured out. Because I didnโt, and I know how overwhelming the jump from ME to AI feels.
Why this matters:
1. For industry: If a simple ML model can give 2 weeks warning before a bearing fails, thatโs money saved.
2. For ME students: You donโt need a CS degree to start. You need Python + the right data + the discipline to ship 1 project. Iโm proving it step by step.
Iโm a Mechanical Engineering grad diving into AI/ML, and Iโm documenting the whole journey publicly.
Not because I have it figured out. Because I didnโt, and I know how overwhelming the jump from ME to AI feels.
What Iโm doing right now:
Building a real predictive maintenance project. Using real bearing vibration data to train a model that can tell โnormal vs failingโ equipment. No toy datasets. Code in Python, results on GitHub.