Embedded Files
10 Academy profile site
10 Academy profile site
Aron Sinkie
Addis Ababa, Ethiopia
Addis Ababa Science and Technology University(2019-2022)
MSc. (Computer Engineering)
MSc. (Computer Engineering)
Email: birliesinkie2010@gmail.com
Programming Language
Programming Language
Python
JavaScript
SQL
C++
Generative AI Tools
Generative AI Tools
Fintunnnig LLM
RAG pipeline
Langchain
Promt Design
Evaluation
Automation Tools
Automation Tools
Docker
GitHub Actions
MLFlow, DVC, CML
Unit-Testing
WandB
General Skill-Sets
General Skill-Sets
ETL Pipeline Building
GAN
Data Visualization
ML Model Life Cycle Management
About me
A skilled Generative AI Engineer and Cybersecurity Analyst proficient in computer engineering, database-design, OOPs, algorithms, data structures, and Python. Strong expertise in building generative AI models, fine-tuning LLMs, RAG systems, prompt engineering, and AI Chat-bot. Proficient in ETL pipeline, data processing, and visualization. Actively expanding knowledge in Quantum Computing, AI, IoT. Committed to staying ahead through self-study and hands-on learning.
Education
- 10 Academy
(Generative AI/Machine Learning Engineering/Data Engineering/WEB 3 Engineering)
(Generative AI/Machine Learning Engineering/Data Engineering/WEB 3 Engineering)
Relevant Courses
Relevant Courses
Large language model and design LLM apps(GPT-4, Mistral 7b, Stable Diffusion)
RAG Pipeline
Machine Learning Pipeline Development
Data Engineering Principles and large Scale Implementations
WEB 3 dAPP Development
Statistical Reasoning
- Addis Ababa Science and Technology University (2019-2022)
MSc. (Computer Engineering)
MSc. (Computer Engineering)
Relevant Courses
Relevant Courses
Machine Learning and Intelligent System
Computer Vision
Computational and Applied Mathematics
Random and Stochastic Processes
Advanced Distributed System
High Performance Computing
Parallel Computing Architecture and Programming
- Addis Ababa Science and Technology University (2015-2021)
BSc. (Computer Engineering)
BSc. (Computer Engineering)
Relevant Courses
Relevant Courses
Introduction to AI
Embedded System
Computer Architecture and Organization
Relational Database Management
Digital Signal Processing
Probability and Statistics
Data Structures & Algorithm Design
Image Processing & Pattern Recognition
Computational Method
Discrete Mathematics
Operating system Principles
Build a local language spell checker for the Gurage language and contributing to chat-bot development as measured by enhancing accuracy by 30% by utilizing a novel distance calculator .
Developed an optimized and lightweight Network Intrusion Detection System (NIDS) to detect network attacks and improved memory consumption by 20% by customizing the tabular dataloader to a numpy dataloader of the fast.ai framework.
served as an Anti-fraud Analyst, utilizing machine learning techniques to identify and prevent fraud as measured by successfully reducing fraudulent transactions by 25% by developing and implementing a custom fraud detection model.
Reduced analysis time by 30% by optimizing the use of the X86 Instruction set in malware analysis, resulting in improved efficiency and faster threat detection.
Redesigned the user interface for the internal and open-source tools used in malware analysis, enhancing usability and simplifying workflows for the security team.
Implemented advanced static and dynamic analysis techniques for desktop app malware and mobile app malware, improving detection rates by 25% and
enhancing overall threat mitigation capabilities.
Improved malware analysis effectiveness by reverse engineering malware on Windows systems, providing valuable insights into malicious behaviors and aiding in
the development of more robust security measures.
Increased malware detection accuracy by 20% through the integration of cutting-edge threat intelligence feeds, ensuring proactive identification of emerging malware
threats.
- Embedded System Developer Intern
lead the successful development of hydroponic systems, measured by their efficiency in nutrient delivery, plant growth rates, and overall crop yields. This involved designing and implementing innovative hydroponic setups, optimizing nutrient solutions, and leveraging automation to monitor and control environmental factors for optimal plant growth.
Successfully completed smart beehive projects, measured by their impact on beekeeping practices, hive health monitoring, and honey production. This involved integrating sensors and IoT technologies into beehives, enabling real-time monitoring of hive conditions, such as temperature, humidity, and bee behavior, to support informed decision-making and enhance bee colony management.
Projects
Lizzy AI, an early-stage Israeli startup, is spearheading the advancement of contract AI by developing a next-generation Retrieval-Augmented Generation (RAG) system for Contract Q&A. The objective is to build a fully autonomous artificial contract lawyer, revolutionizing the legal industry. This project focuses on creating an advanced legal contract assistant powered by state-of-the-art RAG technology, empowering legal professionals with cutting-edge AI solutions.
The core focus is on providing key services, namely Automatic Prompt Generation, Automatic Evaluation Data Generation, and Prompt Testing and Ranking. These services collectively contribute to the overarching goal of making advanced AI capabilities more accessible and user-friendly. In a rapidly evolving digital landscape where speed and accuracy are paramount, PromptlyTech’s solutions cater to the dynamic needs of businesses, offering a competitive edge in leveraging LLMs for various applications.
A city traffic department wants to collect traffic data using swarm drones from a number of locations in the city and use the data collected for improving traffic flow in the city and for a number of other undisclosed projects. I was responsible for creating a data warehouse that will host the vehicle trajectory data extracted by analyzing footage taken by swarm drones and static roadside cameras.
This project develops a Redash chatbot add-on that leverages advanced language models and natural language interaction for extracting actionable insights from YouTube data. It bridges the gap between non-technical users and complex SQL queries, democratizing data exploration. By integrating large language models into Redash, the add-on enables team members to interact with Redash through natural language conversations. It utilizes advanced language models to understand user queries, generating SQL queries that are executed against YouTube data. The add-on seamlessly integrates with Redash, presenting visualizations within the chatbot interface and enhancing data analytics capabilities.
Page updated
Google Sites
Report abuse