Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I am Shivani Teotia, currently pursuing a Bachelor of Technology in Computer Science with a specialization in Artificial Intelligence and Machine Learning from Bennett University, Noida. I have hands-on experience developing scalable applications using modern technologies such as Next.js, TypeScript, and Docker, and I enjoy building solutions that involve real-time collaboration and blockchain integration. I am passionate about AI and software development, demonstrated through various projects like a real-time collaborative canvas app and a Python code autocompleter fine-tuned from CodeLlama. I also have a strong interest in cybersecurity and have been recognized nationally as a finalist in the KAVACH Cyber Security Hackathon for creating AI-based multilingual spam detection solutions.…I am Shivani Teotia, currently pursuing a Bachelor of Technology in Computer Science with a specialization in Artificial Intelligence and Machine Learning from Bennett University, Noida. I have hands-on experience developing scalable applications using modern technologies such as Next.js, TypeScript, and Docker, and I enjoy building solutions that involve real-time collaboration and blockchain integration. I am passionate about AI and software development, demonstrated through various projects like a real-time collaborative canvas app and a Python code autocompleter fine-tuned from CodeLlama. I also have a strong interest in cybersecurity and have been recognized nationally as a finalist in the KAVACH Cyber Security Hackathon for creating AI-based multilingual spam detection solutions.

Shivani Teotia

Developer, Web Developer, Programmer, +2





I am Shivani Teotia, currently pursuing a Bachelor of Technology in Computer Science with a specialization in Artificial Intelligence and Machine Learning from Bennett University, Noida. I have hands-on experience developing scalable applications using modern technologies such as Next.js, TypeScript, and Docker, and I enjoy building solutions that involve real-time collaboration and blockchain integration. I am passionate about AI and software development, demonstrated through various projects like a real-time collaborative canvas app and a Python code autocompleter fine-tuned from CodeLlama. I also have a strong interest in cybersecurity and have been recognized nationally as a finalist in the KAVACH Cyber Security Hackathon for creating AI-based multilingual spam detection solutions.…I am Shivani Teotia, currently pursuing a Bachelor of Technology in Computer Science with a specialization in Artificial Intelligence and Machine Learning from Bennett University, Noida. I have hands-on experience developing scalable applications using modern technologies such as Next.js, TypeScript, and Docker, and I enjoy building solutions that involve real-time collaboration and blockchain integration. I am passionate about AI and software development, demonstrated through various projects like a real-time collaborative canvas app and a Python code autocompleter fine-tuned from CodeLlama. I also have a strong interest in cybersecurity and have been recognized nationally as a finalist in the KAVACH Cyber Security Hackathon for creating AI-based multilingual spam detection solutions.

Available to hire

I am Shivani Teotia, currently pursuing a Bachelor of Technology in Computer Science with a specialization in Artificial Intelligence and Machine Learning from Bennett University, Noida. I have hands-on experience developing scalable applications using modern technologies such as Next.js, TypeScript, and Docker, and I enjoy building solutions that involve real-time collaboration and blockchain integration.

I am passionate about AI and software development, demonstrated through various projects like a real-time collaborative canvas app and a Python code autocompleter fine-tuned from CodeLlama. I also have a strong interest in cybersecurity and have been recognized nationally as a finalist in the KAVACH Cyber Security Hackathon for creating AI-based multilingual spam detection solutions.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Language

English

Advanced

Hindi

Advanced

Work Experience

Developer at EasyETH

January 1, 2025 - Present

Developed a wallet adapter interface supporting injected wallets such as MetaMask using the Wagmi library. Enabled Ethereum transfers on both testnet and mainnet, dynamically selecting the correct network based on user context. Designed a user-friendly interface for initiating transfers and monitoring wallet connection status in real time.

Developer at Temulate

January 1, 2025 - Present

Developed a real-time collaborative canvas application with drawing and chat features using a monorepo architecture. Built scalable services including an HTTP API with Express and a custom WebSocket server, both in TypeScript. Managed Redis queues to handle deletion of inactive rooms and maintained drawing state in memory. Implemented cookie-based authentication shared across frontend and backend, using a shared PostgreSQL database for persistent data. All services were dockerized for independent deployment.

AI Engineer / Researcher at Python Code Auto Completer Project

November 30, 2023 - August 5, 2025

Fine-tuned Meta's CodeLlama-7B model on a specialized dataset for Python code auto-completion using QLoRA to optimize for limited hardware. Created training prompts from docstrings and code, applied 4-bit quantization to reduce memory footprint, and achieved measurable improvements validated by ROUGE-L and F1 scores. Deployed the model on Hugging Face Hub and tested with the Transformers pipeline.

Hackathon Finalist Developer at KAVACH Cyber Security Hackathon

August 31, 2023 - August 5, 2025

Developed a multilingual mobile app using Flutter to detect and block spam calls, messages, and emails with AI/ML-based classifiers leveraging BERT. Supported four major Indian languages. Built a web dashboard for real-time display of crowdsourced and AI-flagged blacklisted numbers and emails. Trained models using metadata, behavioral patterns, and user feedback, ranking in the top 3 teams nationally.