Frankline Oyolo
Lead Machine Learning Researcher with demonstrated experience Engineering ML Pipelines and leading development of scalable data-driven systems. Current work in Applied 𝔻eep Neural Networks & ℝeinforcement Learning.
Work Experience
Manulife Financial Corporation
Summer Analyst : Asia Data Office
- Heading development of Pixstory's Retrieval Augmented Generation (RAG) based Conversational Search, accurately answering user queries by grounding responses in a verified knowledge base and generating nuanced, de-opinionated answers.
- Mitigating hallucination issues in large language models and addressing misinformation by ensuring reliable information retrieval through rigorous source vetting and verification. The retrieval process is re-ranked based on the level of trustworthiness of the sources.
- Public beta launch upcoming, featuring an immersive and user-friendly interface. Stay tuned for updates!
(May 2023 - Present)
The University of Hong Kong
Artificial Intelligence Research Assistant
- Led development of conflict resolution dialogue agents for DARPA’s Civil Sanctuary Program, as an Applied Researcher with the Natural Language group (CUTE LAB NAME) at USC-ISI.
- Designed and deployed a scalable chatbot for mitigating toxic behavior on social media through non-violent communication strategies, including prompting GPT-3’s advanced transformer-based (the model that drives ChatGPT!) generation capabilities.
- Moderated over 15k multilingual posts with 85% true positive rate in French and German on popular subreddits.
- You can find the published research paper here, which will presented at NAACL 2024.
(January 2022 - July 2023)
The University of Hong Kong
Data Engineering Research Assistant
- Contributed in developing the conversational AI for Lumin.ai as a Smart Scheduler & Sales Accelerator to enhance customer interactions and optimize the sales funnel for 5+ product and franchise owners.
- Advocated for efficient documentation using Markdown and managed customer interactions for multiple clients.
(May 2022 - August 2022)
African Leadership Group
Software Development Engineer - Machine Learning
- Boosted user engagement on Discovery Plus (D+) as part of the ”Personalization and Recommendations” team.
- Employed Apache Spark and Airflow to build efficient ETL pipelines in a distributed environment, and executed A/B tests to validate and refine new recommendation features for the D+ platform.
- Led the development of server-side ad-insertion SDKs for Android and Web, driving increased ad revenue for D+ sports content in Europe.
(September 2020 - August 2021)
URBN (HKSTP Ideation Stage II)
Co-Founder, CTO and Lead Machine Learning Operations
- Spearheaded the research, design, and development of MiDAS, a micro-service-based system serving personalized ads to millions of users on OTT platforms.
- Designed and trained Computer Vision models with TensorFlow, deployed on Kubernetes to enable dynamic server-side ad-insertion for 40+ channels, boosting revenue for U.S. publishers.
- Developed a tool to automate the dataset annotation process, by leveraging SCTE-35 markers in HLS and DASH streams.
- Leveraged multi-step recurrent networks on time series data for ad break forecasting. This resulted in a 30% reduction in resource consumption and network calls to the SSAI backend.
- Member of the senior development team at AdSparx that was acquired by Warner Bros Discovery in September 2020.
(March 2019 - September 2020)
Moto Health Technologies Inc.
Lead Machine Learning Engineer
- Led and mentored a team of 30 for catering Machine Learning applications with on-Edge capabilities, for Embedded Linux platforms.
- Developed a crowd flow analysis product using face and voice recognition for event registration, optimized for on-chip GPUs to showcase AI-on-Edge computing capabilities.
- Worked on Pose estimation for detection and tracking of human body pose to create an immersive experience for users for a video gaming platform.
(May 2018 - January 2019)
Publications
Comparative Evaluation of Pre-Trained Models, VGG16 And EfficientNetB7, for Brain Tumour Segmentation; Baseline for Clinical Implementation
Frankline Oyolo
The 2024 AWS Re:Invent Conference Las Vegas
Theoretical Astrophysics Research on How The Big Crunch Builds to the Big Bounce Theory
Frankline Oyolo
International Collegiate Journal of Science
Alternative Proposal to the Big Bang’s Singularity Theory: the Hartle-hawking State
Frankline Oyolo
International Collegiate Journal of Science
Detecting Gravitational Waves with Precision: Evolution of Gravitational Wave Detectors
Frankline Oyolo
International Collegiate Journal of Science
How the Bystander Effect on Security Robots can Hamper Law Enforcement
Frankline Oyolo
International Collegiate Journal of Science
Curriculum
The University of Hong Kong
Bachelor of Engineering in Data Science and Engineering
Specialization in Machine Learning & Algorithms
Coursework Highlights
- Advanced Natural Lanuage Processing - An Advanced course, helping develop expertise in reproduction of state-of-the-art NLP research and technical publication writing.
- Advanced Computer Vision - An Advanced course providing an overview on the traditional and Deep Learning techniques for Computer Vision.
Holberton School
Software Engineering
Distinction grade
Coursework Highlights
- Advanced Natural Lanuage Processing - An Advanced course, helping develop expertise in reproduction of state-of-the-art NLP research and technical publication writing.
- Advanced Computer Vision - An Advanced course providing an overview on the traditional and Deep Learning techniques for Computer Vision.
Kapsabet High School
Software Engineering
Distinction grade
Coursework Highlights
- Advanced Natural Lanuage Processing - An Advanced course, helping develop expertise in reproduction of state-of-the-art NLP research and technical publication writing.
- Advanced Computer Vision - An Advanced course providing an overview on the traditional and Deep Learning techniques for Computer Vision.
Mathematics for Machine Learning
- Advisory Board Member - August 2017 to July 2018
Applied Machine Learning
- Vice Chairman - July 2016 to September 2017]
Advanced Data Structures, RSA and Quantum Algorithm
- Vice Chairman - July 2016 to September 2017]
Algorithms
- Vice Chairman - July 2016 to September 2017]
Algorithms I and II
- Vice Chairman - July 2016 to September 2017]
Internet of Things (IoT)
- Vice Chairman - July 2016 to September 2017]
Applied Machine Learning
- Vice Chairman - July 2016 to September 2017]
Introduction to Machine Learning
- Vice Chairman - July 2016 to September 2017]
Machine Learning Operations (MLOps)
- Vice Chairman - July 2016 to September 2017]
Technical Skills
Programming Languages
Python / C / C++ / Javascript / Kotlin / R /
Machine Learning
Tensorflow / DeepLab 3+/ PyTorch / MXNet / PySpark / OpenCV / CUDA / Librosa / NLTK / SpaCy / OpenNLP / TensorRT / TFLite
Databases
SQL / NoSQL / Hadoop / Redis / MongoDB / InfluxDB / ElasticSearch / Kafka / Splunk / Datadog
Frameworks
Docker / Kubernetes / Airflow / Tableau / Django / Flask
Cloud Platforms
Amazon Web Services (AWS) / Google Cloud Platform (GCP) / Microsoft Azure
Frontend
HTML / CSS / Javascript / ReactJS / NodeJS / Bootstrap / VueJS
Testimonials
“
Frankline helped me bring up my product, kazi social , to light by applying agile Android development practices , cybersecurity and Web Development using CSS/SSLlabs/PHP/Shopify rising our client outreach by 75%. As a senior product managerr, He Executed Transition of Android development from Java to Kotlin using Android studio to increase device support by 50% and boost user engagement/app to 500 users/day.
Mr Gabriel Wambua
Senior Brand Manager - Safaricom
“
Frankline Possesses the acumen of a ML Engineering team Lead and has been working alongside our CTO to curate best Tech for our company. He Engineers the Full Stack Kotlin Android Development Cycle (ASDLC) Project, Moto Inc., using Android Studio to integrate AWS API, Google Firebase User Metrics API, Payment Gateway APIs, and UI/UX taste using Figma/Adobe/Canva and improve client suppport by 70%
Evans Nomboh
Chief Executive Officer @ Moto Health Technologies Inc.
“
Frankline was the best Data Analyst we ever had. he introduced the concept of data automation using Machine Learning. HeApplied the KNN Model using scikit learn using Google Collab to a 500-person Attendance Dataset to match student E-mail to an accuracy of 96.6% which reduced workload in my office by 40% compared to our initial manual email-student matching.
Edward Opoku
Lead Data scientist - ALGGet In Touch