10 Academy profile site
Amon Kimutai
Nairobi, Kenya
Jomo Kenyatta University of Agriculture and Technology (JKUAT)
BSc. (Control and Instrumentation)
Email: amonkimutai97@gmail.com
Python programming
Pipeline Tools: Kafka, Airflow, Spark
MLOps & CI/CD: DVC, MLFlow, Travis CI
Streamlit
About me
A junior Data Engineer seeking to leverage my experience and ardent desire in building data pipelines that lay groundwork for revealing game-changing insights in the industry. I utilize data to draw value.
Education
- 10 Academy (Jun - Oct 2021)
Machine Learning and Data Science Training
- Jomo Kenyatta University of Agriculture and Technology (2016-2020)
BSc.(Control and Instrumentation)
Work Experience
Data analysis on clients' data requests
Querying data from the MySQL Database using SQL
Splitting and combine files using R Studio.
Performing Data quality checks and cleaning data with Excel.
Data ingestion using UI tools(Data importer and Bulk filter)
Data tasks queue management, and providing satisfactory responses to the clients on their requests.
Work Experience
Instruments, sensors and system maintenance
Calibration; materials' belt speed
Overhauling electric motors
Projects
SmartAd A/B testing
Using A/B testing to assess if there was a significant lift in brand awareness after the Smart Ad advertising company added a new feature to the ad. Majorly, the comparison between the traditional ways for A/B and the Machine Learning approach was noted.
Speech Recognition System
In a group of nine, we worked on a speech recognition system that converted Swahili speech to text. The World Food Program required a system that will be in an intelligent form that collects nutritional information of food bought and sold at markets in two different countries in Africa - Ethiopia and Kenya.
Sales Prediction
Performed six weeks sales prediction for a Pharmaceutical company with many operating stores. The factors that influenced the number of sales included promotions, competition, school and state holidays, seasonality, and locality.
Industry-Casualty
There are frustrations in the industries when drawing business insights from a tubular data because it does not answer the question "what if." In this project, causal graphs were used to explore the features and their relationships, causing breast cancer, and establishing the hidden features.
Speech-to-text data collection
Kafka, Airflow and Spark tools are vital in model enhancement and utilization of streaming data. This project focused in designing and building a robust, large scale, fault tolerant, highly available Kafka cluster that can be used to post a sentence and receive an audio file.