| Award / Rank | Event & Organizer | Prize & Achievement Details |
|---|---|---|
| π₯ 2nd Runner-up | Tredence InfinityAI 2025 (Bengaluru) | Won βΉ2,00,000. Engineered an AI-driven web app based on real-time emotion detection. |
| π₯ 1st Runner-up | IIT Delhi CodeHive Hackathon 2025 | Won βΉ1,00,000. Designed a Web3 decentralized onboarding social app on Hive blockchain. |
| π₯ 1st Runner-up | ITM Gwalior Code Kurukshetra 2025 | Won βΉ35,000. Placed with MomCare, a pregnancy health companion application. |
| π₯ 1st Place | IIT Hyderabad Pathway Hackathon 2025 | Won βΉ15,000. Built stream-processing pipelines using Pathway.com framework. |
| π₯ 1st Place (Campus) | Microsoft Imagine Cup 2025 | Won βΉ5,000. Built MomCare, an end-to-end pregnancy health companion application. |
|
Your DSA grind, on your GitHub. Automatically.
|
Self-hosted, multi-tenant LLM gateway.
|
|
AI Bias Detection Platform for LLMs via Ollama.
|
Reclaim disk space from idle Git repositories.
|
|
AI reading companion for web novels and fanfiction, published on Firefox Add-ons and Edge Add-ons.
|
Serverless view counters: one POST to count, one image URL for the badge.
|
|
Multi-dimensional timetable engine: built for campus scheduling and attendance, generalizing to any system that needs to query and slice complex schedules.
|
See which account, OAuth provider, and MFA method you used on every site.
|
More projects
|
Hospital management platform.
|
VS Code extension to sync spelling dictionaries.
|
|
Real-time sign language detection with in-browser inference.
|
Save points for code, for when git is overkill.
|
|
Decentralized social media onboarding gateway.
|
Real-time emotion classifier using MediaPipe.
|
|
Fill in a form, get a portfolio.
|
A dual-fisheye dashcam for the Pi Compute Module 5.
|
-
π Predicting Severity of Type 2 Diabetic Retinopathy Using Retinal Fundus Images with New Deep Learning Architectures from the Inception and ResNet Families
- Publisher: IEEE Xplore (Conference Paper, May 2025)
- URL: Read on IEEE Xplore
- Abstract Summary: Evaluated Inception-V4, ResNet-V1, and ResNet-V2 CNN models using transfer learning on fundus images from 9,948 patients. The ensemble model achieved 94.21% accuracy and 98.53% AUC, offering early diabetic retinopathy classification for clinical support.
-
π EquiLens CorpusGen v1.0.0 β Gender Bias Audit Corpus
- Publisher: Zenodo (Open Research Corpus, Sept 2025)
- DOI:
10.5281/zenodo.17014104 - URL: View Corpus on Zenodo
- Abstract Summary: Open-source dataset generator for bias auditing in Small Language Models (SLMs) with isolated variable controls.
| Certification Name | Issuer | Credential Verification Link |
|---|---|---|
| βοΈ AWS Certified Cloud Practitioner | Amazon Web Services | Verify Credential |
| βοΈ OCI 2025 Certified DevOps Professional | Oracle University | Verify Credential |
| ποΈ OCI 2025 Certified Architect Associate | Oracle University | Verify Credential |
| π§± OCI 2025 Certified Foundations Associate | Oracle University | Verify Credential |
| π NPTEL Cloud Computing (noc25-cs11) | IIT Kharagpur | Verify Certificate |




