Learn Data Skills
Hongbo Peng

Hongbo Peng

Senior Data Scientist


My Portfolio Highlights

My New Track

Data Scientist

My New Course

Introduction to Python

Data adventurer, fearlessly exploring the vast landscapes of information.

My Work

Take a look at my latest work.


Introduction to R


Introduction to Python


Intermediate Python

DataCamp Course Completion

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

My Work Experience

Where I've interned and worked during my career.

Guardian Life | Jan 2018 - Present

Senior Data Scientist

o Extracted and cleaned raw form-5500 data (from US Department of Labor) to build a database that enabled business department the capability of understanding the performance of each regional sale office and the comparison with competitors on market share, broker commission fee etc.; o Extracted data from company internal data lake and built models for insurance product pricing, provider network optimization and fraud, waste and abuse detection; o Built look-alike models to find potential customers for digital marketing campaigns and also a tracking system on the progress of the campaigns.
Show More

None | Jun 2017 - Dec 2017

Kaggle Competitions Expert

won 3 solo silver medals: o top 2% among 3779 participants in completed competition “Zillow Prize: Zillow’s Home Value Prediction”: built stacking regression models (xgboost, lightgbm and catboost) for improving Zillow’s home valuation model; o top 3% among 1257 participants in completed competition “New York City Taxi Trip Duration”: identified external data sources to combine with training data and employed xgboost and lightgbm to build stacking regression models for predicting the total ride duration of taxi trips in New York City; o ranking 31st among 428 participants in completed competition “Santa Gift Matching Challenge”: build min-cost max-flow network optimization model to maximize the total happiness of both “Santa” and children by paring 1 million child with 1000 gifts (amount of 1000 for each gift) based on the constrain of cost/Happiness matrix of gift/child pair and that twins and triplets of children need the same type of gift; o ranking 34th among 260 participants in completed competition “Text Normalization Challenge - English Language”: develop out-of-the-box approach to convert English text including abbreviations, numbers, currency expressions, measure phrases, addresses or dates etc. from written expressions into appropriate "spoken" forms.

IBM | Aug 2010 - Sep 2014

Research Staff Member

Natural language processing and Machine learning (09/2013-09/2014); Bionanotechnology (08/2010-09/2013) o Developed python/c++ programs and applied machine learning algorithms with natural language processing ; o Wrote MPI and cython codes for parallel computing in distributed and shared memory systems respectively; o Built a Linux server with GPU computing capability for deep learning; o 25 granted patents and 7 pending patent applications in the area of bionanotechnology.

IBM | Oct 2007 - Aug 2010

Postdoctoral Research Fellow

Bionanotechnology o Initiated a research project on DNA sequencing (a new area for IBM) in 2007 from scratch and wrote a scientific proposal that won $2.5 million governmental grant in 2009; o The work eventually earned a 3-year business contract of millions of dollars for IBM in 2010.

My Education

Take a look at my formal education

Doctor of Philosophy - PhD, PhysicsBrown University | 2007
Bachelor of Science - BS, PhysicsUniversity of Science and Technology of China | 2000

Powered by

  • Work
  • Courses
  • Experience
  • Education
  • Create Your Data Portfolio for Free