Prabesh Sharma

Hi, I'm Prabesh Sharma

A Computer Engineer Specializing in AI & ML Solutions.

CV

Education

Bachelor in Computer Engineering

Tribhuvan University, Institute of Engineering (TU, IOE)

2021 - 2025

Focused on software development, artificial intelligence, and machine learning. Relevant Coursework: Data Mining, Microprocessor, Object Oriented Programming, Artificial Intelligence and Theory of Computation

Experience

AI Engineer

Inspiring Lab

Aug 2025 - Present

  • Developed multiple agents for planning, researching, database querying, and presentation creation.
  • Built scalable backend APIs to handle document ingestion, queuing ingestion process with RabbitMQ, deletion, slide content management, and more.
  • Created multimodal agents to search and retrieve specific content across diverse document types
  • Optimized agent performance to reduce processing time without compromising accuracy.
  • Mentoring AI interns on building reinforcement learning based AI agents capable of learning from feedback.

    • Software Engineer-AI

    Skillrank

    Feb 2025 - July 2025

  • Part of the team and contributed to develop an AI-powered SDLC automation tool
  • Engineered a system that generates a complete end-to-end codebase with automated GitHub integration, Supabase setup, and one-click deployment all from a single prompt or instruction.
  • Built LLM workflows to generate architecture, design, and context in one streaming call, reducing cost by 25%
  • Develop Agents for Figma-to-code conversion to streamline the design-to-development pipeline.
  • Automate backend deployment on one click.

    • AI/ML Engineer Intern

    Scala Technologies

    Oct 2024 - Feb 2025

  • Developed models for brain tumor segmentation, including 2D and 3D architectures.
  • Designed and implemented route optimization solutions
  • Built AI-powered chatbots for various use cases.
  • Worked with large language models (LLMs) for Specific domains.
  • Created customizable bots for automation and user interaction.
  • Contributed to demand forecasting models to enhance decision-making
  • Designed and implemented model for hatespeech detection.

    • Skills

    • Programming: Python, SQL
    • Data Analysis: Pandas, NumPy, Matplotlib, Seahorn
    • Machine Learning: Scikit-learn, TensorFlow, PyTorch
    • Deep Learning: LSTM, Transformers, CNNs
    • LLMS and RAG
    • Computer Vision
    • Backend Development (Django)
    • API Development (DRF)
    • Git, GitHub, Docker
    Download CV

    Projects

    Brain Tumor Segmentation

    Unsual Activity Detections

    Designed a real-time anomaly (Fighting, Accidents, Robbery, Vandalism, Explosions,etc 13 Anomaly) detection system using fine-tuned Vision Transformers. Integrated WebRTC for live streaming, WebSockets for real-time communication, and Telegram API for alerts.

    Python Django Vision Transformers WebRTC WebSockets Flutter Telegram API
    GitHub GitHub Demo Research Article
    Crop Recommendation System

    IMAGE CAPTIONING

    Developed an image captioning system using Python, Django, and deep learning, integrating Transformer architectures like BERT, GPT, and T5 alongside VGG16 and LSTM. Enhanced feature extraction, caption generation, and NLP refinement for seamless user interaction and accurate, coherent captions.

    Python Transformer VGG16 CNN NLP & LSTM Django
    GitHub GitHub
    Brain Tumor Segmentation

    ONLINE COURSE RECOMMENDATION SYSTEM

    Developed an Online Course Recommendation System using Python, Django, and NLP techniques like Tf-Idf and cosine similarity, with curated data for accurate, tailored recommendations with my minor project team.

    Python NLP Django
    GitHub GitHub
    PDF Question-Answering System

    PDF Question-Answering System

    Streamlit-based app leveraging Weaviate for vectorized Q&A over the uploaded documents. PDF Question Answering is the system where user can upload the pdf and ask anything about the pdfs

    Python Streamlit Weaviate RAG Langchain
    GitHub GitHub
    Brain Tumor Segmentation

    EMAIL GENERATOR FOR THE JOB POST

    Developed an automated cold email generator using LangChain, Llama 3.1, and ChromaDB for precise job matching and tailored applications

    Python llama3.1 Chromadb Langchain
    GitHub GitHub
    Brain Tumor Segmentation

    COTTON PLANT DISEASES CLASSIFICATION

    Developed a cotton plant disease classification system using CNN and Django, enabling precise image analysis for early detection and improved crop management.

    Python CNN Django
    GitHub GitHub
    Brain Tumor Segmentation

    STUDY ROOM

    Built Studybuddy, a website using HTML, CSS, JavaScript, and Django, enabling users to create/join rooms, connect, chat, and share knowledge in their fields of interest.

    Python Django Frontend Concepts
    GitHub GitHub

    My Blog Posts

    Medium Medium

    Image Captioning Using Transformer

    After numerous attempts with RNNs, GRUs, and LSTMs to generate captions for images...

    May 30, 2024 12 min read
    Read on Medium →
    Dev.to Medium

    Machine Learning Algorithms

    Machine Learning is said as a subset of artificial intelligence that is mainly concerned with the developmen...

    Dec 9, 2023 18 min read
    Read on Medium →
    Hashnode Medium

    Database and SQL for Data Science With Python

    SQL A language used for relational Databases Query data...

    Nov 2, 2023 12 min read
    Read on Medium →

    Contact

    You can find me on these platforms: