Ishika Ishani

Owned by Ishika Ishani
Last updated: May 23, 2024
3 min read

Hi, I'm Ishika Ishani, a second year B.Tech student in Computer Science at Cochin University of Science and Technology, and pursuing a B.S Degree in Data Science from IIT Madras.

As a Consultant at Napmar Automations, I specialized in AI automation services using platforms like Make, particularly in developing and deploying Voiceflow-based chatbots. Additionally, as a Frontend Developer at Realyt Estate, I contributed to the startup's growth by designing and implementing React-based landing pages, ensuring impactful digital first impressions.

Languages:  JavaScript, Java, C++(OOPS), Python, SQL
Frameworks: TailwindCSS, BootStrap, ReactJS, NodeJS, ExpressJS, Django, Flask
Tools: GIT, GitHub, Postman, Linux, Bash , Notion , SQLWorkbench
Data Analytics : Pandas, NumPy , Seaborn , Matplotlib
Database : MySQL, MongoDB
ML: ScikitLearn, Jupyter Notebooks, Supervised & Unsupervised Models

Outside of academics and professional endeavors, I actively engage in tech blogging and social media, with a significant following and engagement. I also serve as the Chapter Lead of the Tpg Kerala community, supporting women and non-binary individuals in accessing opportunities in Web3 and Blockchain. Furthermore, I hold the position of Joint Secretary at ACES, CUSAT, where I contribute to fostering a vibrant tech community.

Mini Projects Link : https://porfolio-ruby-theta.vercel.app/
Github : https://github.com/ishani-1255
LinkedIn : https://www.linkedin.com/in/ishika-ishani/
Email : ishikaishani97@gmail.com
HashNode : https://ishikaishani.hashnode.dev/

My Ideas for Project :

Since we need to wrap up the ML model into an API, we need to host it on a server and make use of FastAPI to get the actual api and integrate it with the frontend part. While testing the model, I saw "use via api" option in Gradio, I tried integrating the api with a React application.but faced "React UnhandledSchemeError - "node:buffer" is not handled by plugins" error though i am trying to find a solution to that, but if we used the default way of implementation, I would use django for the backend with fast api along with react frontend.

For the frontend I will :

  1. Create a new React application , Install necessary packages like axios, react-google-login, and react-router-dom and implement Google authentication using react-google-login. 

      2. Develop a Login component for user authentication , create a Chat component to handle user input and display responses.

      3. Use useState and useContext for managing user authentication state ,use axios to make API calls to the FastAPI backend whenever a user sends a message.

      4. Implement routing using react-router-dom to navigate between login and chat components.

      5. Design a user-friendly interface with a side panel for past conversations and an input area for new messages.

For the backend I will :

  1. Create a new Django project and app for user authentication. Install Django, Django REST Framework, and other necessary packages.

      2. Set up django-allauth and configure Google OAuth2 authentication. Define models for user profiles and any other necessary data structures.

      3. Develop views and serializers for user authentication and profile management. Create REST API endpoints for login, signup, and token management.

      4. Run migrations to set up the database schema.

For FastAPI Development I will :

      1. Create a new FastAPI application , install FastAPI, Uvicorn, and any other necessary packages.

      2. Define Pydantic models for request and response schemas. and load and integrate the machine learning model for inference within the FastAPI application.

     3. Create API endpoints to handle chatbot queries

     4. Configure CORS to allow frontend requests and use Uvicorn to run the FastAPI server locally for development and testing.