OLAWALE IBRAHIM
Machine Learning Engineer
London,ENG
Summary
Globally recognized talent in the field of digital technology, endorsed by the UK Home Office for exceptional promise. Over four years of professional experience as a machine learning engineer, consistently developing cutting-edge solutions for complex problems. Passion for development, optimization, and deployment of machine learning models in production systems.
Overview
2
2
years of professional experience
Work History
Machine Learning Engineer
Earth Science Analytics
02.2022 - 12.2022
- Developed and productionised ml backend service via a django backend for image classification, semantic segmentation, and rock type detection using SOTA models YOLO, mask RCNN, ResNet
- Deployed training and prediction APIs and endpoints for 3D CNN architectures AttentionUnet, TransUnet, Unet+++, ResUnet for self supervised and unsupervised machine learning denoising of seismic images
- Developed 3D rock property prediction machine learning models using 1D CNN architectures
- Productionized pretrained segmentation, denoising models in EarthNET and EarthVision for clients
- Designed and built RESTful APIs using django and django restframework enabling key functionalities
- Developed advanced data science techniques for improved well log interpretation, to validate workflows and improve interpretation confidence for SMEs
- Effectively addressed and resolved bugs and issues in EarthNet and its modules via django
- Created training materials and presentations, showcasing EarthNet capabilities to all users/clients
- Supported the research team in achieving KPI targets and project milestones
Machine Learning Developer
dGB Earth Sciences
03.2021 - 02.2022
- Built and integrated a PyTorch machine learning plugin into C++ software, creating more support and improvement in workflow efficiency
- Developed machine learning segmentation models for interpreting various seismic and well-log data, using 2D and 3D CNN models in PyTorch and Tensorflow, and classical ML algorithms in scikit learn
- Key open source contributor to the dgbpy GitHub repository
- Pioneered innovative machine learning and data-driven approaches, leading to 100% improvement in well lithology interpretation accuracy
- Continuously improved existing software-based workflows, resulting in increase in user satisfaction
- Created comprehensive tutorials and open-source technical content, reaching thousands of users
Education
B.Tech - Applied Geophysics
Federal University of Technology Akure
Skills
LLMs & Fine tuning
Machine learning
Deep learning
Software Development
Agile Methodologies
Computer Vision
Model Deployment
Prompt Engineering
CI/CD
MLOps
NLP
Llama
GPT
T5
Sloth
Hugging Face
Transformers
Langchain
Llama Index
Vector databases
qdrant
MongoDB
pytorch
tensorflow
huggingface
django
fastapi
flask
postgresql
mysql
redis
neo4j
zenml
mlflow
cometML
Opik
Azure
AWS
S3
Sagemaker
ECR
cloudformation
docker
git
jira
Tech Stack
Llama, GPT, T5, Sloth, Hugging Face, Transformers, Langchain, Llama Index, Vector databases (qdrant), MongoDB, pytorch, tensorflow, huggingface, django, fastapi, flask, postgresql, mysql, mongodb, redis, neo4j, qdrant, zenml, mlflow, cometML, Opik, Azure, AWS, S3, Sagemaker, ECR, cloudformation, docker, git, jira
Awards
- Global Talent of Exceptional Promise, UK Home Office, 2024
- 1st Place (Team), SEG NVIDIA Seismic Facies Classification Competition, 2021
- 2nd Place, NAPE Annual International Conferences: 'Convolutional Neural Networks for Salt Bodies Mapping using Machine Learning', 2021
- Top Position, Steam Optimization Challenge using Machine Learning for Enhanced Oil & Gas Recovery, 2021
- €2000 Prize and Certificate, FORCE Machine Learning Lithology Prediction competition - Developed the state-of-the-art model for lithofacies prediction, 2020
- Best Poster Award, NAPE Annual International Conferences - 'Estimation of missing petrophysical data from well logs using a Machine Learning Based Bagging Approach', 2020
Publications
- A Novel Approach to Train Self-Supervised Seismic Denoising Dnn Architectures, EarthDoc, FirstBreak, 2022
- Convolutional Neural Networks for Salt Bodies Mapping using Machine Learning, NAPE, 2021
- Estimation of Missing Petrophysical Data using a Machine Learning Based Bagging Approach, NAPE, 2020
Work Availability
monday
tuesday
wednesday
thursday
friday
saturday
sunday
morning
afternoon
evening
swipe to browse
Work Preference
Work Type
Full TimeContract WorkLocation Preference
On-SiteRemoteHybridImportant To Me
Career advancementTimeline
Machine Learning Engineer
Earth Science Analytics
02.2022 - 12.2022
Machine Learning Developer
dGB Earth Sciences
03.2021 - 02.2022
B.Tech - Applied Geophysics
Federal University of Technology Akure