Highlights
- Pro
Block or report Mandy1200
Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.
You must be logged in to block users.
Contact GitHub support about this user’s behavior. Learn more about reporting abuse.
Report abuseA Full-Stack Developer & AI Engineer crafting intelligent, scalable, and user-centric applications.
I transform complex problems into elegant solutions, blending machine learning with robust backend and frontend technologies. Currently exploring the frontier of Generative AI with LangChain, RAG, and local LLMs.
Tech: React.js, TypeScript, Tailwind CSS, Vite , Node.js, Supabase (PostgreSQL), LiveKit, WebRTC, Websockets, Zustand
- Architected a production-grade multi-tenant AI voice platform using React, TypeScript, Supabase, and LiveKit, enabling real-time agent orchestration, workflow automation, and analytics.
- Designed a scalable Supabase PostgreSQL backend with RLS, RBAC, authentication, and real-time synchronization for secure multi-tenant deployments.
- Built a no-code workflow engine for AI-driven voice pipelines (Voice Call → AI Processing → Retry Logic → SMS → CRM Integration → Webhooks) , integrating LLMs, STT, TTS, SMS, CRM, and webhooks.
- Reduced voice response latency by 75% (≈1000 ms → ≈250 ms) through streaming optimization and efficient LLM/STT/TTS pipeline design.
- Developed secure REST APIs and reusable TypeScript components, improving maintainability, onboarding, and overall development velocity.
Tech: React.js, PHP, Composer, jQuery, Automation
- Built production-ready modules for an artisan marketplace, including dashboards, workflow automation, and product management systems.
- Developed reusable UI components in React, improving development efficiency and user experience.
- Automated backend processes using PHP + Composer, reducing manual workload by 40%.
- Strengthened digital presence for artisans, enhancing product visibility and storytelling.
- 🧠 AI & Machine Learning: Architecting and deploying end-to-end ML pipelines, building intelligent agents with LangChain/LangGraph, and running local inference with Ollama.
- 💻 Full-Stack Development: Building performant, responsive web applications with the Next.js/React ecosystem and powering them with robust Node.js/Express backends.
- 🛠️ Data-Driven Apps & Viz: Transforming raw data into interactive applications and insightful dashboards using Streamlit, Plotly, and Power BI.
- ☁️ DevOps & Tooling: Proficient in Linux/Bash, version control with Git, and containerization concepts, ensuring smooth development and deployment cycles.
|
A research-grade multi-agent runtime that safely executes AI-generated code using AST sandboxing, multi-LLM consensus, LangGraph orchestration, runtime anomaly detection, and reinforcement learning feedback. Python LangGraph PyTorch Groq SQLite |
A production-grade Hybrid RAG assistant combining FAISS, BM25, cross-encoder reranking, stateful travel planning, Supabase cloud sync, and optimized semantic retrieval for outbound Indian travelers. FastAPI PyTorch FAISS Supabase LangGraph RAGs Transformers |
|
A robotics-powered agricultural platform using Edge Impulse machine learning models and embedded hardware for precision pollination. Arduino Edge Impulse C++ IoT Machine Learning |
A wearable AI communication system translating Morse inputs into multilingual speech using embedded hardware and machine learning. Python IoT Machine Learning React |
|
An autonomous disaster-response platform integrating GPS navigation, obstacle detection, and intelligent routing for emergency supply delivery. IoT Python GPS React |
A collaborative development environment featuring live collaboration, AI-assisted coding, project workspaces, and real-time synchronization. Next.js React Node.js WebSockets |
|
An explainable AI platform for burnout prediction using SHAP-based model interpretability and interactive visual analytics. Streamlit SHAP Plotly scikit-learn |
An NLP-powered resume screening platform using BERT embeddings and FAISS similarity search for candidate matching and ranking. BERT FAISS Streamlit Python |
|
AWS Credit Program Selection Selected at the prestigious PITCH ‘n’ WIN event at IIT Bhubaneswar for innovation and potential. |
|
|
Multi-Hackathon Champion (5x Winner) Dominating the innovation circuit with wins at: IIIT D3 Fest, Silicon SparkUp Summit, Silicon Techtronics 2024, Utkarsh 2024 (StartUp Odisha), and the MSME IDEA Hackathon 4.0. |
|
|
Dual Hackathon Winner Secured top positions for its collaborative features at the IIIT BBSR Invento Expo Advaita & ELYSIUM 2025 KODLAMA (AI Hackathon). |
I'm always excited to connect with new people and explore new opportunities. I'm currently open to:
- AI/ML Research Internships (Remote or In-Person)
- Open-Source Contributions in ML, GenAI, or Frontend tools
- Freelance/Contract Roles focused on AI-driven web applications
- Collaborating on innovative tech projects
Made with ❤️ and a lot of ☕
Pinned Loading
If the problem persists, check the GitHub status page or contact support.