
Nikunj Sehgal
@nikunjsehgal
ML engineering candidate whose GNDU research benchmarked 10 CNNs and achieved 98.02% accuracy.
What I'm looking for
At the Department of Electronics Technology, GNDU Amritsar, I benchmarked 10 CNN architectures for potato leaf disease detection using a unified transfer learning framework. MobileNetV3-Large achieved 98.02% accuracy and an AUC of 0.9987.
I extended the research to five-class turmeric disease detection, where ResNet-50 reached 95.41% accuracy. Grad-CAM confirmed biologically meaningful features, and I documented a preprocessing mismatch that caused accuracy drops in DenseNet-201 and Xception.
I authored an IEEE-format paper and built a CNN benchmark suite that automates training and evaluation, generates Grad-CAM heatmaps, and produces result tables, confusion matrices, and ROC curves.
My other projects include a Java POS system with inventory tracking and sales management, and a Power BI business analytics dashboard connected to a SQL-to-Excel data pipeline. I’m seeking a software or ML engineering role where I can apply research-backed engineering in a production environment.
Experience
Work history, roles, and key accomplishments
Deep Learning Research Intern
Guru Nanak Dev University
Jan 2026 - May 2026 (4 months)
Benchmarked 10 CNN architectures for agricultural disease detection, achieving 98.02% accuracy with MobileNetV3-Large. Authored an IEEE-format paper and presented findings.
Education
Degrees, certifications, and relevant coursework
Guru Nanak Dev University
Bachelor of Technology, Electronics and Communication Engineering
2022 - 2026
Grade: 6.46 CGPA
Pursued a Bachelor of Technology in Electronics and Communication Engineering, graduating in June 2026 with a CGPA of 6.46.
Bhavan's SL Public School
Class XII
Grade: 82%
Completed Class XII with 82%.
The Senior Study School-II
Class X
Grade: 80%
Completed Class X with 80%.
Guru Nanak Dev University
Deep Learning Research Training, Deep Learning
Completed a 5-month deep learning research training program in May 2026.
Tech stack
Software and tools used professionally
Availability
Location
Authorized to work in
Salary expectations
Job categories
Interested in hiring Nikunj?
You can contact Nikunj and 600K+ other talented remote workers on Himalayas.
Message NikunjGet matched with your dream remote job
Sign up now and join over 1,000,000+ remote workers who receive personalized job alerts, curated job matches, and more for free!
