Learn Data Skills
Beta
Peter Olapade

Peter Olapade

Data Scientist

UT Austin

Technologies

DataCamp Course Completion

Take a look at all the courses I’ve completed on DataCamp.

Courses

Introduction to Python
Intermediate Python
Data Manipulation with pandas
Python Data Science Toolbox (Part 1)

My Work Experience

Where I've interned and worked during my career.

Pinnacle | May 2020 - Present

Principal Data Scientist

Developed and deployed containerized production-grade ML and statistical models for cloud-based predictive maintenance analytic solutions. • Developed an automated outlier detection and data cleaning algorithm for health-monitoring sensor data. Built time series forecasting models to predict equipment failure using the pre-preprocessed sensor data. • Implemented an autoencoder using conv1D and LSTM to detect anomalies in time series data. • Built Bayesian inference models to quantify the uncertainty of parametric models for predicting asset useful life. • Performed text pre-processing and topic modeling using Gensim’s LDA model to identify top keywords (the most frequently repaired items) in each topic. • Designed and developed a graph network algorithm to model throughput through complex industrial plants. • Profiled ML calculation, identified and refactored bottleneck increasing performance by more than 20%. • Utilized Azure Kubernetes Services and KEDA to auto-scale docker containerized ML calculation. • Monitored ML microservice applications using Opentelemetry with metrics routed to the Grafana dashboard.
Show More

Halliburton | Jul 2019 - May 2020

Principal Data Scientist

Researched and developed data analytic solutions for controlling hydraulic fracture growth. • Performed data cleanings such as data alignment, filtering using low pass filter, outlier removal, and extracting ML features from time series using Fourier Transform. • Built and trained ML models for predicting dominant fracture length in real time.

Halliburton | Jan 2017 - Jun 2019

Senior Data Scientist

Designed and implemented data interpretation models and algorithms using downhole-acquired logging data. Extensively use Python to perform digital signal processing, data cleaning, machine learning, and optimization. • Built, trained, and deployed an ML model using an ensemble method and non-linear optimization, improving prediction by > 10% compared to the classic curve fitting method. • Developed a DL model for predicting fluid properties from sensor data increasing accuracy by more than 20% compared to the classic linear-based model. • Built and trained ML models to predict fluid properties not measured downhole and optimum sampling location. • Maintained SQL database for storing optical sensor data.

Halliburton | Jul 2016 - Jan 2017

Senior Software Developer

Created algorithms for estimating reservoir fluid contamination and fluid composition in real time. • Delivered re-write of 3 major MATLAB-based calculation engines in Python, enabling integration into C++ applications. • Produced data outlier detection and smoothing algorithm for downhole measured sensor data. • Invented automated real-time QC for predicting the composition of reservoir fluid utilizing k-NN and NN. • Implemented NN model for classifying the type of reservoir fluid, allowing proper completion designs.

Schlumberger | Jan 2013 - Apr 2016

Software Engineer

Developed compositional fluid model, heat transfer model, and multiphase flow model for a commercial multiphase flow simulator (PIPESIM) that is widely used in the oil and gas industry. My accomplishments include: -- Developed a flow-regime-independent multiphase flow model using nonlinear constrained optimization. -- Implemented convective heat model in PIPESIM that significantly improved simulation results especially for viscous flow. -- Improved fluid component characterization, flash calculation, salinity analysis, and phase envelope prediction in PIPESIM. -- Developed wrapper for calling Third Party compositional engine APIs in PIPESIM. -- Developed REST API as a web service for communicating with PIPESIM computational engine.

Schlumberger | Nov 2011 - Dec 2012

Post Doctoral Research Scientist

-- Developed an algorithm to determine optimized parameters for a flow- regime-independent multiphase flow model using nonlinear constraint optimization

University of Texas at Austin | Aug 2007 - Nov 2011

Graduate Research Assistant

-- Analyzed capillary pressure experimental data and developed a correlation for capillary pressure-saturation relation. -- Developed transient multiphase flow models to study effective water (by- product) removal in low-temperature Proton-Exchange Membrane Fuel Cell (PEMFC). -- Developed steady-state multiphase flow model to study parameters for optimum membrane hydration of high-temperature PEMFC

Los Alamos National Laboratory | Mar 2010 - Dec 2010

Graduate Student Researcher

Analyzed optical data and developed an algorithm to predict water saturation in PEMFC

University of Delaware | Aug 2005 - Aug 2007

Research Assistant

-- Investigated pair-wise interactions between droplet fluids in both Newtonian and non-Newtonian fluid flow. -- Developed a front-tracking Direct Numerical Simulation (DNS) approach for studying the multiphase flow behavior and rheological properties of concentration emulsions

My Education

Take a look at my formal education

PhD, Mechanical EngineeringThe University of Texas at Austin | 2011
MSc, Mechanical EngineeringUniversity of Delaware | 2007
BSc, Mechanical EngineeringObafemi Awolowo University | 2002

About Me

Peter Olapade

Peter hasn't filled in a bio text

Powered by

  • Courses
  • Experience
  • Education
  • About Me
  • Create Your Data Portfolio for Free