6-day current streak·15-day longest streak
Ajay Soni Applied AI · Data Science · Computer Vision · Practical ML Systems BCA Data Science student building practical machine learning, computer vision, and data-driven systems. Focused on turning…
Ajay Soni
Applied AI · Data Science · Computer Vision · Practical ML Systems
BCA Data Science student building practical machine learning, computer vision, and data-driven systems.
Focused on turning experiments, notebooks, and models into usable tools, structured workflows, and real-world applications.
---
Profile
I work on applied AI and data science projects where the goal is not only model training, but also building systems that are understandable, usable, and documented.
My current focus areas are machine learning workflows, computer vision pipelines, notebook-based experimentation, data science tools, and edge-focused AI ideas.
---
Featured Public Repositories
VXN-RAMNet
GPS-free visual route memory research system using video frames, EfficientNetB0 embeddings, similarity matching, and backtracking branch-graph logic.
AI-DS-100
Structured AI and Data Science project lab with implemented notebook bundles across basic, intermediate, and advanced learning levels.
---
Core Areas
Machine Learning
Model training, evaluation, classification, regression, and workflow automation.
Computer Vision
Visual route memory, CNN pipelines, image classification, and edge-focused vision systems.
Data Science Tools
EDA tools, AutoML workflows, notebooks, dashboards, and structured project labs.
---
Technical Stack
Languages
AI, Data Science and Notebooks
Web, Tools and Deployment
---
Current Learning Direction
Mathematics for AI/ML
Python and model implementation
Computer vision systems
Real-world deployment workflows
---
GitHub Stats
---
Learning by building. Improving by shipping. Documenting the process.
-
VXN-RAMNet ★ PINNED
GPS-free visual route memory research system that learns routes from video frames, builds backtracking branch graphs with EfficientNetB0 embeddings, and classifies query routes using similarity evidence.
Jupyter Notebook ★ 67 1mo agoExplain → -
AI-DS-100 ★ PINNED
A structured AI and Data Science project lab with 66 implemented notebook bundles across basic, intermediate, and advanced levels, including datasets, PDFs, and concise project descriptions.
★ 70 26d agoExplain → -
AutoEDA-Tool
An R Shiny exploratory data analysis app for CSV upload, summaries, missing values, outlier detection, distributions, correlations, visual plots, and interactive dataset review workflows.
R ★ 6 1mo agoExplain → -
Waste-Segmentation-CNN
A smart waste classification and optimization prototype with a CNN-based Flask interface, bin status simulation, route logic, RAPTOR flowchart, project report, and presentation files.
Jupyter Notebook ★ 5 25d agoExplain → -
Weather-Sense-ML
A C + Python weather condition classifier with a Windows GUI, Random Forest model, saved encoder/model files, training notebook, balanced dataset, prediction script, and input history.
C ★ 5 1mo agoExplain → -
Num-Guess-C
A Windows-based C number guessing game using Win32 controls, hints, restart flow, attempt tracking, score updates, and a simple graphical interface for beginner C practice and GUI logic learning.
C ★ 5 1mo agoExplain → -
AutoML-Tool
A Python Shiny AutoML prototype for dataset upload, preprocessing, encoding, feature generation, regression/classification comparison, model selection, and prediction input workflow.
Python ★ 5 1mo agoExplain → -
PerceptronX-Lib
A custom Python perceptron learning library with regression, binary classification, experimental multiclass support, scaling helpers, gradient descent logic, prediction, and scoring.
Python ★ 5 1mo agoExplain → -
AjaySoni-Dev
Applied Machine Learning and Computer Vision enthusiast focused on building practical, real-world AI systems beyond experiments and demos.
★ 1 21d agoExplain →
No repos match these filters.