Vikrant Kadam
AI & Data Science Student. I love building explainable, real-world ML systems.
- •Machine Learning, Explainable AI (XAI), and full-stack development excite me the most.
- •Competitive programmer, open-source contributor, and researcher.
- •I believe in building things that are useful and explainable, not just accurate.
Here are my socials
Vikrant Kadam
@VikrantKadam028
Open-source contributor and competitive programmer. View repos, PRs, and contribution history.
Vikrant Kadam - He/Him - 3rd
@VIT'28 | Full Stack Web Developer | AI and Emerging Tech Enthusiast
Vishwakarma Institute of Technology
500+ connections
Vikrant Kadam
vikrantkadam28
Rank 2,27,163
1,548
Rating
482
Solved
33.06%
Top
vikrantkk2889@gmail.com
Direct email
Usually replies within a day. Open to opportunities and collaborations.
Do I Wanna Know?Last played
Arctic Monkeys · AM
I love designing and building thoughtful, production-grade applications.
View Project
DEXTER
Build By Team DextersDEXTER is a local-first, agentic application security assessment CLI built around a real Scan → Analyze → Verify → Refine loop instead of single-pass scanning. A ReAct tool-calling agent investigates ambiguous findings before trusting them — reading code, checking secret entropy, and grepping the repo — across 14 integrated tools spanning SAST, secret scanning, dependency CVEs, and authorization-gated recon/DAST, with every tool run sandboxed in a disposable Docker container.
View Project
Credé
Credé is an explainable AI credit-scoring platform with separate portals for borrowers and banks. An XGBoost model predicts creditworthiness from user and financial profile data, with SHAP-based explanations breaking down exactly which factors drove each score. Features include OCR-based document ingestion, a bank-side applicant dashboard with fraud alerts, loan application workflows, and a live risk engine — all served through a Flask API and a React dashboard.
Experiences
- Engineered a full-stack AI content generator, processing 5,000+ API requests/day with optimized multi-turn chat history management and secure JWT authentication
- Developed 10+ production web pages using PHP and MySQL that served over 2,500+ active users, executing asset compression and lazy loading to reduce average load times by 22%
- Collaborated in an Agile development environment, participating in sprint planning and code reviews to deliver 3+ client-facing features on schedule while maintaining cross-browser compatibility across Chrome, Firefox, and Safari
- Engineered and published @vikrant_kadam/regex-simplifier on npm, a modular library that translates plain-English descriptions into production-ready regex patterns
- Designed the package's public API and TypeScript types for developer-friendly integration, along with documentation and usage examples to drive adoption
- Achieved 499+ organic downloads within the first 7 days of launch, growing usage entirely through community discovery with zero paid promotion
Proof Of Work
I live spending time in open source,building real stuff and solving real problems
Pull Requests
All pull requests
Publications & Research
Peer-reviewed work at the intersection of AI, accessibility, and real-world systems.
A Multimodal and Deep Learning-Based Image Captioning System for Visually Impaired Persons with Audio Assistance
Viomesh Singh, Vikrant Kadam, Anushka Pise, Prachi Dudhankar, Onkar Bansode
A real-time image captioning system built for visually impaired users, combining CNN-based computer vision with NLP to generate natural-language scene descriptions, paired with speech-to-text input and text-to-speech output for a fully voice-driven accessibility workflow.
2026 Fourth International Conference on Secure Cyber Computing and Communications (ICSCCC) · Jalandhar, India · 29–31 May 2026
Soily: Crop Recommendation Using Satellite Imagery for Western Maharashtra with Machine Learning Algorithm
Sheela Chinchmalatpure, Prachi Dudhankar, Vikrant Kadam, Onkar Bansode, Anushka Pise
Soily is a crop and fertilizer recommendation web app for Western Maharashtra, fusing SoilGrids soil proxies, Sentinel-2 NDVI, and OpenWeatherMap climate data with a Random Forest classifier — reaching 99.7% validation accuracy for crop prediction and 90.0% for fertilizer prediction, outperforming Logistic Regression, KNN, SVM, and a tabular CNN baseline.
4th International Conference on MAchine inTelligence for Research & Innovations (MAiTRI 2026) · NERIST, Itanagar, Arunachal Pradesh, India · 8–9 May 2026
Skills
I love working with these technologies to build beautiful and functional applications.