6th Semester CSE Student

Pulkit Sardana

AI/ML - Computer Vision - Deep Learning

I build practical machine learning projects around images, restoration, recommendations, and full-stack applications. I am looking for internships where I can learn from strong engineers and contribute to real project work.

Pulkit Sardana
4Projects
1Certificate Earned
1Research Project
AI/ML& Computer Vision

Projects

Built around real implementation work

The first three projects are the main focus of this portfolio: AD²-Restormer, Solvix, and an intelligent recommendation engine.

Project output from Solvix

Computer Vision / Research

Solvix

A solar panel defect recognition system using real annotated images and interpretable visual features.

Key result89.82% crop-level accuracy and 92.57% defect-only accuracy on a 6,493-image Roboflow dataset.
  • YOLO annotation to crop-level classification
  • 13 handcrafted CV features
  • Weighted k-NN with maintenance recommendations
PythonOpenCVScikit-learnNumPyRoboflow
User featuresItem metadataSimilarity score
1
2
3
4

Machine Learning

Intelligent Recommendation Engine

A recommendation workflow combining content-based similarity and collaborative filtering ideas.

Key resultBuilt an end-to-end ML pipeline for feature engineering, similarity scoring, ranking, and evaluation.
  • Content and user-item signals
  • Similarity scoring
  • Ranking-focused evaluation
PythonPandasNumPyScikit-learn
WiseMart catalog itemWiseMart catalog itemWiseMart catalog itemWiseMart catalog item

Full Stack + Recommendations

WiseMart

A React and Flask e-commerce project with authentication, cart flow, checkout, and product recommendations.

Key resultConnected a practical shopping interface with recommendation models for similar and frequently purchased products.
  • JWT auth
  • Cart and checkout flow
  • Recommendation endpoints
ReactFlaskPythonJavaScriptJWT

Certificates

1 Certificate Earned

Building expertise through continuous learning in AI/ML, Computer Vision, Deep Learning, and Software Engineering.

Current Certification

Stanford Online and DeepLearning.AI via Coursera

Machine Learning Specialization

Completed three courses covering supervised learning, advanced learning algorithms, unsupervised learning, recommender systems, and reinforcement learning.

More certifications coming soon.

About

A student portfolio, not a company page

I am still early in my engineering journey, but I care about building things that are understandable, reproducible, and useful.

I started with machine learning fundamentals through coursework and the Stanford Machine Learning Specialization. From there, I began building projects that helped me connect theory with real code.

My strongest interest right now is computer vision. AD²-Restormer helped me think about deep learning architectures, while Solvix taught me how much the dataset, feature design, and evaluation matter.

I am also learning full-stack development so I can build interfaces and APIs around the models I create.

Currently studying

B.E. Computer Science and Engineering at RVITM, Bengaluru, expected graduation in 2027.

What I like exploring

Computer vision, deep learning models, recommendation systems, and the engineering needed to turn experiments into usable projects.

How I learn

I learn best by building: preparing data, writing training code, checking metrics, documenting what worked, and improving the result.

Skills

Tools I use most

Grouped by where they show up in my projects rather than by self-scored percentages.

AI / ML

PyTorchScikit-learnNumPyPandasModel Evaluation

Computer Vision

OpenCVImage RestorationFeature ExtractionTransformers

Software

PythonJavaScriptReactNext.jsFlaskGit

Contact

Open to internships and research opportunities

I am looking for AI/ML, computer vision, software engineering, and research internship opportunities where I can keep learning and build useful things.