Parham Rezaei

• Establishing online classes for questions and answers and solving problems
• Supported student learning objectives through personalized and small group assistance.
• Documented attendance and completed assignments to maintain full class and student records.
• Mentored students through Telegram Group and one-on-one communication.

Implement different Machine Learning microservices using ml server

work with FastAPI

works with text to music model

works with music source separation libraries and try different techniques.

• Authoring a Book on Raspberry Pi Pico
• Reading the Different English Resource and assesses the projects they mentioned with my microcontroller
• I have translated the information obtained from all these books into fluent Persian at certain time intervals and I am trying to create a good Persian text resource for teaching this microcontroller

• Ability to transform raw documents into structured data formats suitable for LLMs Context.
• handling dynamic data inputs and ensuring seamless integration into chat sessions.
• Ability to develop chatbots that can interact with users with real time data provided by them
• Gain Skills in working with Embedding models ,Gradio , Langchain , OpenAI API and qdrant Vector DBs

• Understanding the Way Microsoft AutoGen works
• Work with the tool and try to assess the different scenarios with it
• Orchestrate Multiple LLM based Agents to do complex tasks
• Implement it using local LLM with LM studio and Mistral 7B and llama2 70B
• Comparing the Results between GPT 3.5 turbo API, GPT 4 API, llama2 70B and Mistral7B

• fine-tuned the YOLO v8 model with a Persian license plate dataset, achieving high accuracy in license plate detection and recognition.
• Optimized the final model to run efficiently on low-spec PCs, ensuring accessibility and usability in resource-constrained environments.
• Achieved a significant improvement in detection accuracy, making the system reliable for real-world applications.
• Enabled real-time processing capabilities, allowing for immediate recognition and response

• Understanding and Re-Implementing ESRGAN
• Extracting Assets from three famous video games
• Make Training, Validation and Test datasets by just placing random assets from each game to different sections
• Create and Train a Model based on the Paper and Compare the results with the existing ones