SRM INSTITUTE OF SCIENCE & TECHNOLOGY- RAMAPURAM

Faculty of Science & Humanities

PG Department of Computer Applications

 

EVENT TITLE: One Day Workshop on End-to-End GenAI Project Development: Prompt Design to Streamlit Deployment

DATE: 28/02/2026

TIME/VENUE: 1pm – 4pm Inline Zoom Meet

COORDINATORS: Dr.D.Kanchana, Assistant Professor

GUEST DETAILS: Dr. Shreekant Jere, Generative AI Engineer at Accenture AI, Bangalore

SUMMARY OF THE EVENT

1. Event Overview

​The PG Department of Computer Applications (Faculty of Science and Humanities) hosted a specialized one-day workshop titled “End-to-End GenAI Project Development: Prompt Design to Streamlit Deployment.” The session aimed to provide participants with a roadmap for building production-ready Generative AI applications using industry-standard tools.

​

2. Resource Person Profile

​The session was conducted by Dr. Shreekant Jere, a Generative AI Engineer at Accenture AI, Bangalore. Dr. Jere shared insights from his professional experience in the field, moving beyond basic AI interactions to professional application architecture.

3. Technical Curriculum & Key Concepts

  • ​Foundations of Prompting: Participants learned that a “Prompt” is a specific instruction or input provided to an AI to trigger a desired response. Examples covered text generation, creative writing, and code generation.
  • ​The Quality Evaluation Framework: A core principle discussed was the post-generation assessment. Dr. Jere explained how to refine AI outputs by evaluating:
    • ​Clarity: Effectiveness in communicating features and benefits.
    • ​Appeal: Visual and practical resonance for the user.
    • ​Accuracy: Functional alignment with the initial request.
  • ​The GenAI Tech Stack: The workshop focused on a modern, scalable stack including Python, FastAPI for the backend, Uvicorn as the server, and Streamlit for the frontend.

​

4. Practical Project Highlights

​During the session, two primary projects were demonstrated to showcase the versatility of Large Language Models (LLMs):

  • ​Email Attribute Extractor: * A tool designed to parse raw email text and automatically extract key data points (attributes).
    • ​Implementation involved using the OpenAI API and securing keys via Python .env files to follow security best practices.
  • ​Local Slang Changer Model: * A creative application that converts standard language into local slang.
    • ​This project highlighted the use of FastAPI to handle requests and Streamlit to create a user-friendly interface for real-time interaction.

​

5. Key Takeaways for Participants

  1. ​End-to-End Lifecycle: Participants moved from the “Prompt Design” phase to the “Deployment” phase, understanding how to wrap an AI model into a functional web app.
  2. ​API Integration: Hands-on experience with integrating ChatGPT API keys into local development environments.
  3. ​UI/UX for AI: Learning how the Streamlit library allows developers to build data-rich, interactive frontends with minimal CSS or HTML knowledge.