Digital Craftsman (Developer / Programmer)
I am a Full Stack Developer with strong expertise in Next.js, React.js, Node.js, Express, and Python, along with experience in building APIs and backend systems.
- 🌱 I’m currently learning many things, I believe that everyday is a learning opportunity.
- 💁♂️ Trusted member and Moderator at Once UI
- ❤ Contributing to Open Source.
- 💻 Visit my Portfolio for more details about me.
September 2023 - present | Agentic & Generative AI Engineer
- Built and maintained applications that actually work across browsers, reaching 15k+ monthly users. React components, mobile-first designs, the stuff users see and touch
- Designed backend APIs that don't leak secrets (proper auth, validation) and can handle growth without falling over (MongoDB aggregations, query optimization)
- Shipped forecasting pipelines using Python + scikit-learn. Took business problems (demand prediction, sentiment analysis) and turned them into production ML systems
- Worked through the entire lifecycle — from understanding what customers actually need to deploying it and keeping it running. Got comfortable with both the shiny new code and the mess of production systems
- Got comfortable with ambiguity. Early-stage startup meant wearing multiple hats, shipping fast, and iterating based on real usage
- React
- Next.js
- TypeScript
- JavaScript ES6+
- Redux Toolkit
- Component Architecture
- Server-Side Rendering
- Static Site Generation
- Node.js
- Express.js
- Python
- Django
- Django REST Framework
- RESTful API Design
- GraphQL
- API Gateway
- MongoDB
- PostgreSQL
- SQL Queries
- NoSQL Databases
- Redis
- Query Optimization
- Docker
- Docker Compose
- Containerization
- Container Orchestration
- CI/CD Pipelines
- GitHub Actions
- Large Language Models (LLMs)
- OpenAI API Integration
- GPT Models
- Text Generation
- Prompt Engineering
- LLM Fine-tuning
- Retrieval Augmented Generation (RAG)
- Vector Embeddings
- Semantic Search
- LangChain
- LlamaIndex
- AI Agents
- Agent Orchestration
- LangGraph
- ReAct Pattern
- Tool Integration
- Function Calling
- Agent Frameworks
- Multi-step Reasoning
- Text Classification
- Sentiment Analysis
- Named Entity Recognition (NER)
- Text Preprocessing
- Tokenization
- Word Embeddings
- Transformer Models
- BERT
- Text Summarization
- Question Answering
- NLTK
- spaCy
- ML Models
- Supervised Learning
- Unsupervised Learning
- Classification
- Regression
- Clustering
- Feature Engineering
- Data Preprocessing
- Model Evaluation
- Time Series Forecasting
- Scikit-learn
- TensorFlow & PyTorch
- Pandas, NumPy, Matplotlib, Seaborn
Problem: Enterprise legal teams wasted 15+ hours/week on manual contract tracking, review workflows, and compliance checks.
Solution: Built production-grade CLM platform with Django backend (PostgreSQL + Redis + Celery) handling concurrent 10+ users processing 100+ contracts monthly. Async task queue reducing analysis from 40s → 2s.
Impact: 95% reduction in processing time. 40% faster review cycles.
Tech: Next.js, React, TypeScript, Django, PostgreSQL, Redis, Celery
Problem: Students & professionals needed realistic handwriting conversion for assignments.
Solution: SaaS platform converting typed text to handwriting with 99%+ accuracy. Built React canvas engine, Django backend with Stripe integration managing 500+ subscriptions.
Impact: 20+ users. 4.8/5 star rating. PDF, PNG, DOC exports.
Tech: React, TypeScript, Django, PostgreSQL, Redis, Docker
Problem: Sports analytics fragmented across multiple platforms.
Solution: Real-time analytics platform ingesting 100k+ data points/day. Redux state management for 50+ concurrent users. PostgreSQL optimization reducing response from 800ms → 150ms.
Impact: 15+ sports organizations. 99.5% uptime SLA. Dashboard loads in <2s.
Tech: React, TypeScript, Vite, Redux Toolkit, Django, PostgreSQL
Problem: Building scalable dashboards requires complex state management and real-time data sync.
Solution: Full-stack analytics dashboard handling 500+ concurrent sessions. Redux Toolkit reducing renders by 70%. MongoDB aggregation generating reports in <500ms.
Impact: 400+ GitHub stars. Production-ready template.
Tech: MongoDB, Express, React, Node.js, Redux Toolkit, Material UI, Nivo Charts
Problem: Unstructured sentiment analysis lacks actionable insights.
Solution: ML pipeline classifying 10k+ reviews/week with 92% accuracy. BERT fine-tuning improving F1 score from 0.78 → 0.88.
Impact: Deployed in 3 production systems.
Tech: Python, NLP, BERT, Transformers, Machine Learning
Problem: Retail demand forecasting accuracy directly impacts inventory & revenue. Standard models achieved 15-20% MAPE error.
Solution: Ensemble time-series model (ARIMA + XGBoost) achieving 8.2% MAPE on 1,000+ stores. Feature engineering extracting 50+ temporal features.
Impact: Improved forecast accuracy from 15-20% MAPE to 8.2% MAPE.
Tech: Python, Time Series, XGBoost, Scikit-learn, Pandas