AVAILABLE FOR AI/ML & Full-Stack ROLES
Brijesh
Pramanik
|
SCROLL TO EXPLORE
0%
Risk model accuracy
0+
DSA problems solved
0+
Open-source contributions
0 mo
Full-stack ML internship
01 · Profile
About
B.Tech Computer Science graduate with 6 months of AI/ML internship experience in a startup environment. Certified in Microsoft AI-900 and AI-300. Skilled in Python, Machine Learning, Full-Stack Development, Data Analysis, and API Development, with hands-on experience building and deploying end-to-end ML solutions, including medical image analysis. Strong problem-solving and analytical skills with a quick-learning mindset.
02 · Experience
Where I've worked
Full-Stack ML Intern
Mind Webs Ventures · Kolkata
July 2025 — Dec 2025
Bank Transaction Forecasting System
- Developed an XGBoost-based forecasting model achieving 87% accuracy and a 23% RMSE improvement on 50K+ transaction records.
- Improved model performance and reduced training time by 40% using feature engineering and Optuna hyperparameter tuning.
- Built and deployed a FastAPI microservice with real-time prediction capability and an automated ETL pipeline.
- Designed a scalable architecture handling 1,000+ daily API requests, enabling efficient inference and monitoring.
Bank Statement Risk Analysis System
- Developed a machine-learning credit risk scoring system analyzing $2.8M+ in transactions with 92% accuracy.
- Automated loan risk evaluation using 50+ financial indicators, reducing underwriting time significantly.
- Detected hidden financial risks — irregular payments and undisclosed debts — via transaction pattern analysis.
- Built a dynamic risk-scoring model using statistical techniques for real-time financial insight.
PythonPandasNumPy
XGBoostOptunaFastAPI
Scikit-learnMatplotlib
03 · Projects
Selected work
StudyStacks
2026
RAG-powered
OCR pipeline
- Developed a centralized platform to organize handwritten notes, PYQs, and syllabus documents by branch, semester, and subject code.
- Built a RAG-powered AI Assistant that answers academic queries using relevant study materials as context, with source attribution.
- Implemented multi-turn conversational support, enabling follow-up and contextual questions using chat history.
- Developed an automated PDF ingestion and OCR pipeline to extract content from scanned and handwritten study materials.
- Designed a responsive web interface with dark/light theme support for improved accessibility and UX.
React.jsJavaScriptPythonRAGLLMOCRREST APIsPDF Processing
AI-Powered Dental
Anomaly Detection
Anomaly Detection
Jan – Jun 2025
89% accuracy
93% F1
60.03% IoU
- Deep learning model for teeth detection and dental disease classification on panoramic X-rays.
- Used YOLOv7, CNN, ERFNet & U-Net for image segmentation and object detection.
- Applied morphological ops (erosion, dilation) to handle overlapping teeth and refine detections.
- Implemented a majority-voting algorithm for issue classification, outperforming existing models.
TensorFlowPyTorchOpenCVYOLOv7U-Net
Farmly
Multipurpose Platform
Multipurpose Platform
Jan – Apr 2024
2 modules
- Full-stack platform with two modules: Farmer Solution and an E-commerce Site.
- Farmer Solution helps farmers optimize practices and connect directly with buyers.
- E-commerce module lets buyers purchase agricultural products straight from farmers.
JavaScriptHTML5CSSPHPSQLAPI Integration
04 · Skills
Toolkit
Programming
PythonSQLC++JavaScriptHTMLCSS
Data Science & Analytics
PandasNumPyScikit-learnXGBoostFeature EngineeringData PreprocessingData VizARIMASARIMAProphet
AI & Machine Learning
PyTorchTensorFlowPrompt EngineeringGenerative AI / LLM
API & Deployment
FastAPIREST APIsDocker
Tools & Platforms
GitGitHubJupyterPostmanPower BIExcelVS CodeGoogle Colab
05 · Certifications
Verified credentials
Microsoft Certified
Active
Machine Learning Operations
Engineer Associate
✓ Online Verifiable
Microsoft Certified
Fundamentals
Azure AI Fundamentals
✓ Online Verifiable
06 · Education
Academic background
Government College of Engineering & Textile Technology, Berhampore
Bachelor of Technology — Computer Science & Engineering
CGPA 8.81
2025
Nischintapur R.D. High School, Nischintapur
Higher Secondary — WBCHSE
90.60%
2020
07 · Achievements & Leadership
Beyond the resume
Achievements
- 200+ contributions to open-source projects.
- Mentored 200+ students across different tech stacks.
- Selected for the 6th National Science Congress, West Bengal.
- Solved 400+ DSA questions on LeetCode, GfG Practice, and Coding Ninjas.
- All India Rank 170 in CodeKaze (2023), organized by Coding Ninjas.
Positions of Responsibility
- Founder & Co-Lead, Code Nexus — college coding club.
- AI/ML Lead, MLSA-GCETTB (Microsoft Learn Student Ambassadors).
- Student Expert, Postman API Fundamentals.
08 · Resume
Full resume
Brijesh Pramanik — Resume
PDF · Updated 2026
09 · Contact
Let's build something with data & models.
Open to AI/ML engineering and data science roles. The fastest way to reach me is email — feel free to also connect on LinkedIn or check out my code on GitHub.