Robert John
GDE (ML & GCP), Arm Ambassador, Edge Impulse Expert
Lagos, Nigeria
Actions
Robert has worked with databases and developed enterprise software for over a decade. He currently works with all things data, from engineering to visualization and Machine Learning. He recently started working with embedded systems and TinyML and is eager to see the widespread adoption of embedded systems in Africa.
In the past, Robert worked as a software engineer building systems for finance, telecommunications, and utilities/energy vending.
Robert looks forward to a time when Africans will design and build their own embedded systems with Machine Learning capabilities.
He contributes to the developer ecosystem by speaking at events, hosting meetups and code labs, and tutorials. He also writes articles on Hackster.io, Medium, and his personal blog about data visualization, ML, and Google Cloud Platform (GCP). He is a mentor at the Google for Startups Africa Accelerator programme, and is a co-organizer of the GDG Cloud developer community in Kaduna, Nigeria.
Links
Area of Expertise
Topics
Machine Learning with tf.estimator
tf.estimator is the new recommended approach to developing and deploying machine learning models in TensorFlow. We will look at going from low-level APIs to tf.estimator.
Understand the AI/ML Offerings from Google
Google has a wide range of tools available for research, prototyping, and development of ML models and solutions. These range from Actions on Google and ML Kit on the one hand, to AutoML and TensorFlow on the other hand. Find out which solution is suitable for your level of experience and what resources you need to get started.
Machine Learning with GCP
The Google Cloud Platform provides an extensive range of tools for individuals interested in Machine Learning. There are API endpoints for software developers who have no idea what Machine Learning is, but need to roll out solutions that include ML features. There are solutions for those who know what ML is but need to roll out a quick-and-dirty proof-of-concept. And then there are solutions for data scientists and engineers who are working with large datasets to build and deploy custom solutions.
Machine Learning on Tiny Edge Devices
Deploying ML models on mobile devices has become common place, but how about deploying them to tiny microcontrollers that can run on a battery? This workshop will teach participants how to process sound files and train a model, then optimize the models for deployment using TensorFlow Lite for Microcontrollers, and then finally show them how to deploy the model in a feat of embedded programming. This session will require a working knowledge of the C and Python programming languages.
Getting TensorFlow Models into Production
A lot of attention is paid to the research and training of Machine Learning models, but very little attention is paid to deployment. In this session, we take a look at what it takes to get a model developed using tf.estimator or tf.keras into production.
Considerations for Deploying ML Models on Edge Devices
What do you need to keep in mind when training ML models that will be deployed on constrained devices such as phones, Raspberry Pis, or even Arduinos? This session goes into the considerations and the code. snippets required for these.
Considerations for Deploying ML Models on Edge Devices
What do you need to keep in mind when training ML models that will be deployed to phones, Raspberry Pis, or even Arduinos? This session will cover the things that are required in order to successfully train and deploy such models.
Build a Data Startup on GCP
We built a startup that collects, stores, and analyses millions of tweets by making use of a lot of features of GCP. Find out how we make use of Cloud Machine Learning Engine, Pub/Sub, BigQuery, Firebase and Firebase Authentication, Cloud Functions, and lots more.
BigQuery - A Managed Database Instance for Data Warehousing and More
What do you do with all of the data that you collect? How are you analyzing your data? The world is going big on predictive analytics on tabular data. How can BigQuery help you with all these?
Baby Steps to Machine Learning with TensorFlow
There is a lot of buzz about Machine Learning, but how exactly do you get started. How do you get data and train a model? If you would like to know how to get started, then this is the session to attend.
Kigali Devfest 2023 Sessionize Event
DevFest 2022 Lagos Sessionize Event
GDG Kenya DevFest 2022 Sessionize Event
IndabaX Nigeria 2022
Activity Detection on Microcontrollers
Computer Vision with Separable Convolutions
DevFest Kigali 2022
TinyML - A Case Study of the Elephant Edge Collar
I introduced the audience to the problem of animal conservation, and the quest to introduce intelligent solutions using embedded systems in the guise of an elephant collar. I then discussed the possibilities based on the hardware, and then provided an example of how one of those solutions could be built using TensorFlow Lite for Microcontrollers.
IndabaX Namibia 2022
Tiny Machine Learning
The talk was given to attendees of the Deep Learning IndabaX which took place at the Namibia University of Science and Technology, and highlighted how ML is used on embedded devices to aid with animal conservation in Sub-Saharan Africa, and how interested individuals could get started using simple microcontrollers and some sensor data. My session commences around 56 minutes into the provided link.
Google Machine Learning Bootcamp
Paving the Way for Your Success in Machine Learning
This talk was given to participants at the ML Bootcamp organized by Google Developers and Gebeya. My talk centered around the soft skills needed to be successful in the workplace.
2022 Google I/O Extended, Kaduna
Build Your First IoT Application
This an IoT presentation to participants at the I/O Extended event organized by GDG Cloud kaduna. The presentation was designed to show the architecture of an IoT product, and an example of building an internet-connected lamp that can be connected using both a physical button and from a cloud dashboard. I made use of an Arduino microcontroller with an onboard WiFi radio to illustrate this. A how-to for the actual project is available https://www.hackster.io/robert-thas/internet-controlled-light-switch-c4381e
2022 Google I/O Extended, Aba
What Goes Into An IoT Application
This an IoT presentation to participants at the I/O Extended event organized by GDG Calabar. The presentation was designed to show the architecture of an IoT product, and an example of building an internet-connected lamp that can be connected using both a physical button and from a cloud dashboard. I made use of an Arduino microcontroller with an onboard WiFi radio to illustrate this. A how-to for the actual project is available https://www.hackster.io/robert-thas/internet-controlled-light-switch-c4381e
TensorFlow User Groups, Azerbaijan
There Is More Out There!
A presentation about TFLite for Microcontrollers to the TensorFlow User Group Azerbaijan.
DevFest Conakry 2019 Sessionize Event
DevFest 2019 Lagos Sessionize Event
DevFest Southern Africa 2019 Sessionize Event
Google DevFest 2019 - Mauritius Sessionize Event
Robert John
GDE (ML & GCP), Arm Ambassador, Edge Impulse Expert
Lagos, Nigeria
Links
Actions
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top