I have worked with Eka Software Solution as Vice President – Product Engineering. I have 15+ years of experience. My learning and Capstone journey were amazing. We were 5 team members in the Capstone Group.
One of our team members is doing a project in this area already. It is a very initial stage, though. He had a good domain understanding and had first-hand information on the real challenges and the market scope in this area, so we decided to take up this topic.
Out Topic was – Predicting early identification of corrosion in Oil & Gas Pressure Systems using AI & ML techniques.
We wanted to implement all of the learning from the PGP – AIFL program. We were able to use the application of Supervised Learning Techniques (Models like XGBoost, Bagging Classifier, Gradient Boosting Classifier, AdaBoost Classifier, Decision Tree Classifier, and Random Forest Classifier were recommended for classification of corrosion severity)
Apart from this, concepts of Data gathering, data pre-processing, data processing, EDA, and PCA were also studied and explored.
The key highlight / Learning outcome from the capstone project was –
- Application of AI and ML concepts to solve practical business problems
- It gave very good insights on all the phases (ideation to deployment) involved in addressing the business problem end-to-end
- Team composition and sizing
- How to arrive at Revenue models and ROI calculations
- Coming up with a Conference paper was a good learning experience.
- Learnt about a new domain (pressure systems) as part of literature study
- Peer groups presentations provided good and concise insights to AI/ML application in various industries and problem statements
My overall experience was good. It was challenging to collaborate within the group due to different priorities, industry experience, diversity in experience, different interest levels. But putting it all together with the learning experience, it was a good journey overall.
The overall AIFL program journey was good. Throughout the program, we went through the technicalities involved in the AI/ML space, the various types of models, etc.
The capstone was more about doing a startup, sourcing data, people, revenue, and ROI. I enjoyed the learning sessions.
The capstone project could have been a little longer due to the number of details required to be gathered, understood, and shaped into a working startup model.
Lastly, I would like to Thank Priya Bohra for being an Excellent Facilitator. She has done a fabulous job throughout the program. She had been very prompt and always ensured that all the queries get a timely addressed.
Excellent Videos by Dr. Abhinanda Sarkar and Dr. Kumar Muthuraman, and some of the complex topics were explained with great examples and made it look easy.
A Special Thanks To Mr. Anwesh and Dr. Narayan for the mentorship during the Capstone program
And will not forget to mention the review of the capstone projects by Prof Mukesh Roa was excellent and did provide a lot of insights.
