Fahad Ahmed
Lead ML Engineer | Product Manager @
Omdena Inc.
Data Scientist, Machine Learning Engineer and aspiring researcher having a Master's degree in Data Science and a Master's degree in Computer Science.
Having a 3+ years in AI/ML and data analytics and 10+ years of software development experience using C/C++, Java, Python, JavaScript etc.
Led teams implement and integrate analytics and AI in their organizations in order to improve their processes and gain insights to assist in making better decisions.
Portfolio
CervAI - Diagnosing Cervical Cancer using Computer Vision
Lead ML Engineer
Pathology is essential for cancer diagnosis and requires to perform cervical
cancer screening by expert pathologists. There is a severe shortage of experts,
with only 8 pathologists for 15 million Zimbabweans. On average, there is only
one pathologist per 1 million people in Africa. As a result, it may take up to 3
months for cervical cancer tests to be reviewed, as pathologists cannot keep up
with the demand.
By using AI and Deep Learning, pathologists will end up screening only the most likely cancerous samples. Therefore the pathologist will thus have to review only about 35% of the slides, significantly reducing the burden of their workload. This will improve the turnaround time and patient outcomes for the 20,000 women in Zimbabwe whose test samples go to pathologists each year.
By using AI and Deep Learning, pathologists will end up screening only the most likely cancerous samples. Therefore the pathologist will thus have to review only about 35% of the slides, significantly reducing the burden of their workload. This will improve the turnaround time and patient outcomes for the 20,000 women in Zimbabwe whose test samples go to pathologists each year.
Activities:
- Collaborating with the computer vision team (Deep Learning & Computer Vision)
- Researching, building and training deep learning models such as ViTs
[ Project Info ]
Building Concious AI for Sophia Robot - Hanson Robotocs
ML Engineer
Humanity is facing a lot of challenges due to AI. We can already see AI algorithms
suppressing people’s opinions and trying to control us. So to face future challenges,
from bad actors (and AI), we have to build compassionate algorithms. Otherwise, we will
make intelligent machines that kill, fight and suppress each other.
Activities:
- Collaborating in self-supervision team
- Researching,building and training of action recognition deep learning models
[ Project Info ]
Finding Mosquito Species using Computer Vision - VecTech
ML Engineer
Vectech’s current methods for species classification are based on images gathered with MosID, a custom imaging device
designed for consistent, high-quality mosquito imaging.
The goal is to develop a mosquito body part segmentation and identification method,
to help determine what parts of the mosquito are visible and intact in the image.
This enables more advanced computer vision methods, serving as highly valuable prior
information for the CNN, and may be paired with entomological taxonomic information for
species identification. These advanced methods are required for mosquito surveillance
products because captured mosquitoes from the wild often have missing scales, wings,
legs, etc., which sometimes affects whether that mosquito can be accurately identified to
its species.
Activities:
- Collaborated with the machine learning modeling team
- Built, trained and tested a segmentation model using FPN + ResNet50 with a mean IOU score 0.437
[ Project Info ]
Detecting Harmful Video Content and Children Behavior through Computer Vision
Product Manager
Children are more susceptible to acting impulsively and participating in online internet challenges.
Internet challenges can encourage kids to replicate unsafe behaviors to increase user engagement and
materiality on social media. Some of these outrageous challenges have led to severe bodily harm and even death.
To protect children from these types of dangerous ideas and peer pressure, AI is used to filter out this content.
Activities:
- Leadership experience in managing multiple teams from start to completion of the end product
- Engaging customer and collaborators to discussion, meetings, sprints
- Managing the product to be consistent, on-time and within the scope
Tech Domain: Data Visualization, Data Analysis, Web scraping, Web application, Ethical AI, Computer Vision
Industry: Online Safety, Social Media Content Moderation
Predicting Painful Movements For Elbows Using Machine Learning
ML Engineer
Remote assessment with MLI® Elbow from PRECURE helps to prevent strain in the forearm
muscles and therefore reduces the risk of developing the injury. On the one hand,
MLI® Elbow is an intelligent wearable that measures muscle strain by using EMG signals,
which represent the contraction and relaxation of the muscles. On the other hand,
the MLI® technology analyses these EMG signals using risk thresholds based on research
and provides an MLI (Muscle Load Index)
Activities:
- Task leader - EMG signal noise removal and normalization
- Performed feature extraction from RAW EMG signals
- Performed R&D on various EMG Signal processing and visualization libraries
- Using Neurokit to perform EMG Signal processing, noise removal and normalization
- Build and train LSTM model to identify noise in EMG signals and classify them as Noise and Not-Noise to be used in training LSTM model to predict pain levels of muscular activity.
- Performed hyperparameter tuning to improve pain prediction LGBM model performance by 15%
Tech Domain: Data Visualization, Data Analysis, Web scraping, Web application
Industry: Healthcare, Health and Wellbeing
[ Project Info ]
[ Precure]
Pakistan COVID-19 Analysis
ML Engineer
Building computer vision models to early detect the presence of covid through
X-ray images or CT scans and classify the severity of Covid-19 for a patient.
Activities:
- Task Leader - Modeling (Deep Learning & Computer Vision)
- Deep learning model development using Tensorflow and Keras to predict COVID positive or negative in patients using X-ray images.
Tech Domain: Data Visualization, Data Analysis, Web scraping, Web application
Industry: Healthcare, Health and Wellbeing
AI Applied: Reducing the energy crises in Pakistan using Machine Learning
ML Engineer
AI offers great potential to increase the adoption of renewable energy.
To find target sites we need to exclude those that already have electricity. In addition,
those close to the grid were given low priority as they are more likely to receive it
directly in the future. The volume of available and free satellite data is incredible.
There are night-time light images that clearly show towns that have electricity.
Activities:
- Collaborated on collecting data from public databases, newspapers, web pages, and satellite images etc.
- Analyzed the data following a systematic methodology
- Perform exploratory data analysis (EDA)
- and outline an AI-driven solution for Energy Crises.
- Finding solutions to suggest the most necessary locations to install solar power systems which have significant population with no electricity.
Tech Domain: Data Visualization, Data Analysis, Web scraping, Web application
Industry: Energy, Renewable Energy, Solar Energy
Experience
Machine Learning Engineer
Omdena
- Task leader - EMG signal noise removal and normalization
- ML & Deep Learning Modeling ▪️ Build and train LSTM model to identify noise in EMG signals and classify them as Noise and Not-Noise to be used in training LSTM model to predict pain levels of muscular activity.
COVID Analysis in Pakistan - Omdena Pakistan Chapter (OCT - NOV)
Task Leader - Modeling (Deep Learning & Computer Vision)
Deep learning model development using Tensorflow and Keras to predict COVID positive or negative in patients using X-ray images.
COVID Diagnosis using Xray Images App:
https://share.streamlit.io/sfahadahmed/covid19_xray_app/app.py
Predicting Painful Behaviors For Elbow Using Machine Learning - Precure (AUG - OCT)
▪️ Performed feature extraction from RAW EMG signals
▪️ Performed R&D on various EMG Signal processing and visualization libraries
▪️ Using Neurokit to perform EMG Signal processing, noise removal and normalization
▪️ Performed hyperparameter tuning to improve pain prediction LGBM model performance by 15%
Reducing the Energy Crises in Pakistan using Machine Learning- Omdena Pakistan Chapter (JUL - AUG)
▪️ Collaborating on collecting data from public databases, newspapers, web pages, and satellite images etc.▪️ Analyzing the data following a systematic methodology, perform exploratory data analysis and outline an AI-driven solution for Energy Crises.
▪️ Finding solutions to suggest the most necessary locations to install solar power systems which have significant population with no electricity.
Potential sites to install solar powerplants in Pakistan
https://renew-rho.vercel.app/
JUN 2019 - SEP 2021
Data Scientist
Texas Wholesale
- Leading the team for Business Intelligence and Analytics.
- Researching, reporting and advisory on performance and cost efficient solutions and technologies using currently available on-premises and cloud resources.
- Advisor on how the company scales and architects their Data Warehouse and advocate best practices and solutions
- Responsible for the overall architecture and workflows for Business Intelligence and Analytics solutions.
- Architected the on-cloud Data Warehouse solution.
- Design and implementation of hybrid ETL pipeline using on-premises SQL Server Database with Analytics and Integration services (SSIS).
- Implement on-premises to cloud data sync tools which periodically syncs to Amazon AWS S3 and Azure Blob storage.
- Contribute to improving internal data processing tools
- Utilization of statistical and data visualization tools and techniques
- Responsible for monthly Business Intelligence reports and dashboards using Power BI.
- Periodic presentations and storytelling of the gained insights to assist the company executives in making better decisions.
JUN 2019 - SEP 2021
Software Consultant
Acceptto Corporation
- Responsible for front/backend development for an upcoming multi-factor single-sign-on security platform.
- Frontend development using NodeJS, VueJS and CoreUI.
- Backend development using Python and Flask.
DEC 2018 - APR 2019
Senior Software Engineer
Sparco Media LLC
- Responsible for R&D, analysis, design, development and deployment.
- Responsible for overall architecture, development and delivery of various web application long-term projects for Intel OnCue.
- Led development of a B2B project for Sparco.com systems integration with 3rd-party distributor SOAP API's from Tech Data and Ingram Micro using SOA architecture, Java, Spring, Hibernate/JPA, JAXB etc.
- R&D on a surveillance system project using C, C++, Linux, Raspberry pi, parallel computation on multi-core heterogeneous systems using C and OpenCL on Raspberry Pi. Developed core Linux and Windows libraries using C and C++, Integration with .NET CLR using managed C++ as windows services.
- Led the R&D of AssetRabbit.com, a B2C SAAS platform, REST API using Java, Spring MVC/DATA/SECURITY/AOP, JPA/Hibernate, Lucene, ZXing, ExtJS.
- Plugins development for Bugzilla and Chrome.
- Development of various custom scripting tools for Verizon OnCue including data migration and spreadsheet processing tools etc.
July 2008 - October 2018
SOFTWARE CONSULTANT
Alex Solutions
SEP 2018 – OCT 2018
Performed migration TexasWholesale.com ecommerce application to AWS Cloud to achieve high availability, fault tolerance and scalability.
JUL 2018 – AUG 2018
Designed and developed a highly customizable email marketing tool that validates and verifies bulk emails, without actually sending any emails using, only low-level ESMTP/SMTP communication protocols.
OCT 2017 – DEC 2017
Developed the TexasWholesale.com Android and iPhone apps using Ionic 2 and AngularJS 2 frameworks.
JUN 2016 - AUG 2016
Developed the DobeeGo.com Android app using Ionic 2 and AngularJS 2 framework.
JUL 2014 – OCT 2014
Developed the TexasWholesale.com Android app using Ionic 1 and AngularJS 1 frameworks.
July 2014 - OCT 2018
CO-FOUNDER & TECHNICAL LEAD
Xrtus LLC
- Responsible for requirement elicitation with the stakeholders
- R&D, architecture, implementation and delivery of various projects like ConnectWise dashboard, CorTrans logistics dashboard, My-Course UK, NobelCommuniations using Java/JEE, C#/WCF.NET, PHP, jQuery and Bootstrap.
August 2011 – July 2014
SOFTWARE ENGINEER
Greylines Studio
- Led development to delivery of the various flavors of the POI product CityXplorer using J2ME, J2ME Polish, web services using JSON-RPC .NET and C#.
- Improved the overall architecture, reduced redundant code, performance optimization.
- Overhauled data entry processes by introducing new strategies, developed flexible tools using C#.NET for importing, cleaning a reformatting bulk data using Excel sheets and CSV files.
APRIL 2006 – JUNE 2008
MOBILE APPLICATION DEVELOPER
Arbi Industries
- Developed a mobile application for textile industry frequently used calculations for measuring output cloth based on various input parameters.
- Performed R&D on supporting floating-point operations on legacy mobile devices since built-in support was not yet implemented.
- Custom UI widgets for data entry specifically floating point support since no built-in widgets were implemented.
- On-device testing on various legacy Java mobile devices.
2004
WEB DEVELOPER
Arbi Industries
- Responsible for development and deployment of various websites using HTML, CSS, ASP-classic and Dreamweaver.
2001
Education
SZABIST University
Master of Science
Data Science
CGPA: 3.34/4.0
August 2019 - July 2021
University of Karachi
Master of Science
Computer Science
GPA: 4.0/4.0
January 2006 - December 2008
University of Karachi
Post Graduate Diploma
Computer and Information Systems
GPA: 4.0/4.0
January 2005 - December 2005
University of Karachi
Bachelor's Degree
Mathematics
p>GPA: 2.0
January 2002 - December 2004
Skills
Programming Languages & Tools
Awards
I've been awarded a Gold medal for highest achievement in MS (Data Science) in 2021