1  
 2  <h2 align="center">
 3    <img width="25%" alt="Model2Vec logo" src="assets/logo.png"><br/>
 4  </h2>
 5  
 6  This repository is structured as a monorepo, leveraging Git submodules to organize my AI/ML projects and research implementations. The goal is to showcase practical applications and explorations in the rapidly evolving fields of intelligent machine.
 7  
 8  ## 🚀 Domains Covered
 9  
10  This repository categorizes projects into the following AI/ML domains:
11  
12  <!-- 
13  ### ⚙️ Automation
14  
15  Projects focused on automating repetitive tasks, intelligent process automation (IPA), and leveraging AI for operational efficiency.
16  -   [Project 1: Brief Description](#)
17  -   [Project 2: Brief Description](#)
18  
19  ### 💬 Agents
20  
21  Development of conversational AI systems, including rule-based bots, retrieval-based models, and generative models for various applications like customer service, information retrieval, or interactive assistants.
22  -   [Project 1: Brief Description](#)
23  -   [Project 2: Brief Description](#) -->
24  
25  ### 🤖 Agents
26  
27  Development of agent-based for various applications like customer service, information retrieval, or interactive assistants.
28  
29  -   [Pawcha : Conversational agent that tracks and recalls your receipt-based purchases](https://github.com/hfahrudin/pawcha-agent)
30  -   [Nokcha : Your AI Assistant that fuses tabular data and documents into clear fraud insights](https://github.com/hfahrudin/fraud-chatbot)
31  
32  ### 📊 AI/ML Pipeline
33  
34  Tools and frameworks for building, managing, and optimizing end-to-end Machine Learning pipelines, covering data ingestion, preprocessing, model training, evaluation, and deployment.
35  -   [Doc Dup Det: Duplicate detection service powered by embedding model and vector db](https://github.com/hfahrudin/doc_dup_det)
36  -   [Barebone Vector DB: Simplest, minimum-est, implementation of vector DB](https://github.com/hfahrudin/bb_vectordb)
37  -   [QLora-Runner: Fine-tune, quantize, and evaluate LLMs for downstream tasks using QLORA](https://github.com/hfahrudin/qlora_downstream)
38  -   [CTLogs Stream: Rate-limit optimized CT log ingestion via dynamic worker staggering](https://github.com/hfahrudin/ctlogs_stream.git)
39  -   [Dockerize Llamafile: Containerized version of the LlamaFile to making it easy to deploy and manage](https://github.com/hfahrudin/Dockerize-Llamafile)
40  
41  ### 👁️ Computer Vision (CV)
42  
43  Solutions involving the processing, analysis, and understanding of digital images and videos. This may include object detection, image classification, segmentation, and more.
44  -   [Facex: Lightweight, High-Performance Facial Expression Classifier](https://github.com/hfahrudin/facex)
45  -   [Motion Nav: Motion-based PPT Controller](https://github.com/hfahrudin/motion_nav)
46  -   [Face Liveliness Test: POC of webcam-based facial liveness authentication system](https://github.com/hfahrudin/face-liveness-plat.git)
47  
48  ### 📖 Research Implentation
49  
50  Implementations of research papers, focusing on reproducing results, benchmarking models, and exploring novel AI/ML techniques.
51  -   [Meta-Learning Algorithms Implementation](https://github.com/hfahrudin/MetaLearn-MAML_Reptile)
52  -   [Football Defensive Ghosting Agent using Imitation Learning](https://github.com/hfahrudin/Football-MultiAgent-Imitation-Learning)
53  -   [T-Rex Run AI with Deep Reinforcement Learning](https://github.com/hfahrudin/trex-DQN)
54  -   [VigenèreRSAImageLock](https://github.com/hfahrudin/VigenereRSAImageLock)
55  
56  
57  ## 📄 License
58  
59  All projects are licensed under the [MIT License](https://opensource.org/licenses/MIT) - see the `LICENSE` file for details.