10 Academy profile site
Azaria Gebremichael
Addis Ababa, Ethiopia
Addis Ababa Science and Technology University (2017-2021)
BS. (Software Engineering)
Email: azaria.tamrat@gmail.com
Software Engineering
Python, Java, Testing, pytest, unittest, JavaScript, C, C++
Machine Learning
Tensorflow, plotly, Scikit-learn, pandas matplotlib, EDA, Power BI, NLP, Computer Vision
DevOps & MLOps
CI/CD, Docker, Git Kubernetes, CML, DVC, mlflow, Airflow
Data Engineering
SQL, dbt, kafka, redash, superset, AWS RDS
About me
A Software Engineer whose passions lie in building Machine Learning models that translate data points into business insights and bringing them forth through the web. Experience in projects executing data-driven solutions to solve Business and Community problems.
Looking to use my Bachelor of Software Engineering and Business Management to analyze Business problems and implement finer data science solutions. Energetic, and a life-long learner with an active community contribution ready to impact the world using technology.
Education
- Addis Ababa Science and Technology University ( 2017-2021 )
BSc.(Software Engineering)
Relevant coursework:
Object-oriented-programming
Advanced Programming
Software Quality Assurance and Testing
Data Structures and Algorithms
Web design and programming
Software Requirements Engineering
Software Design and Architecture
Systems Programming
Operating Systems
Probability and Statistics
Calculus I
Discrete Mathematics I, II
- Bahirdar University ( 2018-2022 )-- Distance Enrollment
BA.(Business Management)
Relevant coursework:
Introduction to change management
Introduction to Operations and Marketing
Introduction to Financial management
Managing relationships
Managing the change process
Training & Certification
- 10 Academy, Machine Learning
- Tech Camp Ethiopia, MERN stack
- Azure Data Scientist Associate
- Microsoft programming with HTML5, CSS and JavaScript(ES5)
Work Experience
In a team setting, implemented a Text-to-speech engine using approaches including concatenation synthesis and Deep Learning--such as the Tacotron-2 architecture for TechinEye project using Tensorflow,
Layed out flask API used for speech synthesis.
Worked on the frontend application(React PWA).
Co-authored research paper dealing with neural-network approaches for speech synthesis in Amharic.
Created suitable working environment for data science workloads on Azure, run data experiments, trained, managed, optimized &, deployed predictive models & worked on a Deep Learning computer vision project to detect respiratory disorders with accuracy of more than 98%.
As a partner manager in Aiesec for the External Relations team, I helped assess potential partners and influential people to deliver seminars, delivered and attended numerous soft-skills training on communication, networking, collaboration, and leadership.
Featured Projects
Led a group of 7 members to make an Automatic speech recognition system for the Amharic language. Layed out the MLOps pipeline using CML and DVC which had GPU runners integrated within Pull requests to allow easy training on AWS server. Layed out Deployment pipeline utilizing Github actions, Docker hub, and Azure app service. Built and evaluated RNN architectures for Automatic Speech Recognition.
Leveraged the power of transfer learning to evaluate existing architectures and build accurate models to aid the diagnosis of COVID-19 using chest X-ray images.. Multiclass classification models which distinguish between Bacterial pneumonia, viral pneumonia, Tuberculosis, COVID-19, and healthy images. Generated Saliency maps to help identify areas containing salient information about the disorders. Layed out flask API to expose inference model.
Built sales forecasting time-series models using Facebook's prophet, and Random Forest Regressor to predict incoming sales for Rossmann pharmaceuticals for the next 6 weeks. Built analysis dashboard for existing data using power BI and prediction dashboard using plotly JS.
Data Science practice notebooks including malware detection, Breast cancer detection, A/B testing, Animal recognition and more.