UTU Technologies

Work in an agile and iterative way together with the CTO, product manager, research
partners, data scientists and fellow AI/ML and other engineers to devise and implement
impactful and practically feasible AI/ML solutions to build and improve UTU’s trust and
recommendation mechanisms.
● Continuously evaluate and improve the performance of deployed AI/ML solutions.
● Document developed solutions and their results and contribute to white papers, research,
academic publications etc.
● Identify and propose use cases / client sectors to which UTU’s services might be
comparatively easily adapted to.
● Adapt existing AI/ML solutions to new use cases / client sectors of UTU’s services.
● Propose and develop new approaches to improve the performance and effectiveness of our
employed algorithms and mechanisms.
● Review and help to improve algorithms/mechanisms/implementations by data scientists and
fellow AI/ML engineers.
● Inspire and see to the advancement of skills and work style of team members.
● Improve algorithm and code quality as a response to receiving reviews, but also as you find
problems yourself.
● Identification of tools and methodologies which might help improve the AI/ML, product
and/or development process.
● Envision and propose longer term technological goals and approaches, and identify
appropriate existing technological platforms and solutions towards their realisation.

Senior Machine Learning Engineer

Respirosa-US

•	Involved in the data processing, splitting and batch-processing of audio . This is to create the Convolutional Neural Network training dataset.
•	Utilized deep learning technology to train a model to recognize genre of a song when provided with audio information. To achieve this two methods were used:
o	Used a TensorFlow Keras neural network, with a ReLU activation function to explore the numerical features of each song
o	Used FastAI transfer learning technology with the resnet50 model to explore spectrograms created from audio files of each song.
•	Use of a 1-D CNN for Tempo Identification
o	A 10-layer TensorFlow Keras Sequential was used to explore the features on the spectrograms created from the audio files.
•	Carry-out research and development processes on matter concerning deep learning libraries like Keras, PyTorch, TensorFlow, Theano and FastAi . Also efficient as well as cost effective platforms like AWS, PyCharm, Jupyter Notebook and Google Colab. This is to ensure that the AI strategies are fulfilled.

Artificial Intelligence Specialist

PUSH Observer

•	Built a data analytics team, which is tasked with the role of examining raw data of different industries with the purpose of make better decisions in regards to specific client solutions.
•	Coming up with complex statistical models that learn from and scale to petabytes of data.
•	Using of Map-Reduce frameworks such as Pig and Scalding, statistical software such as R and SAS, and scripting languages like Python and Ruby.
•	Writing and interpreting complex SQL queries for standard as well as ad hoc data mining purposes and use of tableau for infographic presentations.
•	Defining metrics; understand A/B testing and statistical measurement of model quality.
•	Understanding and leveraging of crowdsourcing and human computation approaches to data labelling.
•	Using of NewsRadar platform in collection analysing data in the media monitoring department.
•	Developing and managing of new business
•	Developing of new products, as well as overseeing their implementation
•	Participating in the overall expansion of Push Observer
•	Managing  and maintaining of customer relationships with the Company

Data Scientist

Extra Solutions Limited – East African Breweries Limited

•	Extracted and manipulated large quantities of data using tools such as SAS & SQL Server
•	Analysed the customer portfolio to provide insight and guidance to improve processes and increase customer satisfaction.
•	Drove efficiency within Systems and processes.
•	Used statistical methods to help forecast key variables in our Business Planning.
•	Analysed new data sources and incorporating them into databases/analysis suites.
•	Managed and agreed upon priorities and deadlines whilst maintaining a high quality of work.
•	Used the Microsoft office suite to deliver data driven solutions (SQL/excel/access/vba).
•	Used some analytics skills such as verbatim/text analysis to extract insights.

Data Analyst

Kenya National Bureau of Statistics

Scope: Performs a full range of professional and technical duties in support of the KNBS Economic Development Division’s services, projects, and programs; assists with the analysis, coordination, and/or administration of economic development projects and programs; maintains and analyzes economic trends from business retention and expansion program databases; and serves as a resource to other KNBS staff, in providing information and assistance and solving problems.

Economic Development Analyst - EDA Internship

Nairobi Primary School

Scope: Support in the creation and development of mathematical stream policies and programmes to meet the needs of pupils and to improve the level of achievement in the classroom;
To ensure the appropriate levels of development for pupils within class group;
To assume continuity of policy and curriculum promoting equal opportunities for all.

Volunteer Mathematics Teaching Assistant

An Industrious Machine Learning Engineer with 5+ years of experience building successful algorithms and
predictive models in various industries. Highly adept at Recommender systems, Regression and Classification algorithms. A motivated engineer and analyst proficient with various programming languages and the ability to apply ML techniques to solve business problems.

Taplio

Leah Njuguna (MSc.)'s Linkedin Analytics

Leah Njuguna (MSc.)

Want to drive more opportunities from LinkedIn?

Taplio

.css-1jgreyo{background-image:linear-gradient(284deg, rgba(39,43,72,1) 0%, rgba(33,47,183,1) 100%);color:transparent;-webkit-background-clip:text;background-clip:text;}Leah Njuguna (MSc.)'s Linkedin Analytics

Leah Njuguna (MSc.)

Want to drive more opportunities from LinkedIn?

Leah Njuguna (MSc.)'s Linkedin Analytics