Manish Kumar Shah
Full-stack engineer and aspiring data scientist building AI-powered applications across healthcare, agriculture, and enterprise domains. Proficient in React, Next.js, Node.js, and Python with hands-on experience in machine learning, deep learning, and computer vision. Active researcher in medical image analysis (BraTS brain tumor classification and segmentation). Hackathon finalist with a track record of production-grade delivery under pressure.
2
Research manuscripts
2
Hackathon finals
3
Domains
4
Applied projects
About
Full-stack engineer with an applied AI focus
Building AI-powered applications across healthcare, agriculture, and enterprise domains with a balance of product polish and research rigor.
Professional Summary
Full-stack engineer and aspiring data scientist building AI-powered applications across healthcare, agriculture, and enterprise domains. Proficient in React, Next.js, Node.js, and Python with hands-on experience in machine learning, deep learning, and computer vision. Active researcher in medical image analysis (BraTS brain tumor classification and segmentation). Hackathon finalist with a track record of production-grade delivery under pressure. I enjoy shipping production-grade systems under tight timelines while maintaining strong data and research foundations.
Education
B.Tech in CSE, JAIN Deemed to be University (CGPA 8.715/10, expected 2028).
Research Focus
Brain tumor classification and segmentation using BraTS MRI datasets.
Domains
Healthcare, agriculture, and enterprise AI with production-grade delivery.
Developer Journey
- Active researcher in medical imaging and brain tumor analysis on BraTS.
- Built AI-first apps for agriculture, healthcare, and enterprise workflows.
- Hackathon finalist with a track record of delivery under pressure.
Skills
Modern web + applied AI toolkit
Full-stack delivery with hands-on ML, computer vision, and production tooling.
Frontend
Backend
AI/ML
Data Science
Databases
Tools & DevOps
Projects
Flagship builds with measurable impact
Each project balances research depth, engineering execution, and user outcomes.
KRUSHI-SCAN
AI agriculture platform with realtime IoT dashboards and crop disease detection.
Impact
5+ IoT streams monitored live, diagnosis time cut to seconds
- Realtime sensor dashboards for farmers
- Image-based disease detection with confidence scoring
- Regional weather intelligence for crop recommendations
CLARIO
24/7 mental health support platform with AI chatbot and emotion recognition.
Impact
Proactive interventions via realtime emotion sensing
- Facial emotion recognition with DeepFace + OpenCV
- AI chatbot for always-on support
- Mood and sleep analytics dashboards
ONEFLOW
Unified project and task management platform with RBAC and finance modules.
Impact
Centralized workflows with realtime cost visibility
- Role-based access with admin approval flows
- Integrated invoicing, expenses, and PO tracking
- Single-pane project execution
EXPENSIO
Expense management platform with hierarchical approvals and realtime tracking.
Impact
Approval cycles reduced with automated workflows
- Multi-role approvals and audit trails
- Receipt uploads with realtime visibility
- JWT-secured, role-scoped access
Brain Tumor Classification (BraTS)
Multi-class MRI brain tumor classification using deep learning and transfer learning techniques on the BraTS dataset.
Impact
Research-focused medical imaging pipeline for improving diagnostic accuracy and reducing tumor misclassification.
- MRI preprocessing with skull stripping, normalization, and NIfTI slice extraction
- Transfer learning using ResNet and EfficientNet architectures
- Optimized for high classification accuracy on glioma subtypes
- Built visualization and evaluation pipeline for medical imaging analysis
HGG vs LGG Segmentation (BraTS)
Semantic segmentation of glioma tumor regions from multi-modal MRI scans using U-Net based architectures.
Impact
Performance evaluation using Dice Score and Hausdorff Distance on BraTS benchmarks.
- Multi-modal MRI inputs including T1, T2, FLAIR, and T1ce
- Attention-enhanced encoder-decoder segmentation architecture
- Tumor region segmentation with Dice coefficient optimization
- Comparative benchmarking against BraTS evaluation standards
Experience
Research and achievements timeline
Research milestones, hackathon results, and applied engineering impact.
Brain Tumor Classification (BraTS)
2025 - PresentManuscript in preparation
Multi-class MRI tumor classification (Meningioma, Pituitary, Glioma, No Tumor) with CNN + transfer learning.
HGG vs LGG Segmentation (BraTS)
2025 - PresentManuscript in preparation
U-Net and attention models for pixel-level tumor masks with Dice and Hausdorff evaluation.
Hackathon Finalist
2025 - 2026Amalthea IIT-GN 2025, IEEE AVM-IIITM 2026
Finalist teams delivering production-grade systems under tight timelines.
Smart India Hackathon
2025Final Round (College Level)
Selected for SIH 2025 with applied AI solutions for real-world challenges.
Achievements
Hackathons and honors
Finalist placements and competitive research milestones.
Finalist - AVM-IIITM
IEEE Student Branch
2026
Finalist - IIT Gandhinagar ODOO Hackathon
Amalthea (IIT-GN)
2025
Smart India Hackathon Finalist
SIH 2025 (College Level)
2025
Contact
Let us build something exceptional
Share your goals, and I will respond with a focused plan and timeline.