- Blog
- CV generator
- Community
- Documentation
Anh Trinh
Data Scientist at Descartes Underwriting
Paris, France
I'm looking for a challenging and interesting problem to solve. My goal is to be a unicorn, a "Jack of all trades, master of all". I prioritize the breadth of knowledge and I'm ready to go deep if necessary. I like to explore the uncharted water and push myself beyond the limit. I love solving new problems and automating old solutions.
CodersRank Score
What is this?
This represents your current experience. It calculates by analyzing your connected repositories. By measuring your skills by your code, we are creating the ranking, so you can know how good are you comparing to another developers and what you have to improve to be better
Information on how to increase score and ranking details you can find in this blog post.
635.1
CodersRank Rank
Top 2%
Based on:
8 events
Mid
Developer
Python
Top 50
Python
Developer
France
Top 50
Shell
Developer
France
Highest experience points: 0 points,
0 activities in the last year
Python
245.1
exp.
Shell
45.1
exp.
C++
72.3
exp.
Java
52.9
exp.
Jupyter Notebook
17
exp.
Scala
21.4
exp.
JSON
18.6
exp.
R
16.5
exp.
C#
44.6
exp.
F#
8.4
exp.
Haskell
24.5
exp.
Batchfile
18.1
exp.
Rust
21
exp.
List your work history, including any contracts or internships
Descartes Underwriting
Jul 2021 - Present (3 years 5 months)
Data Scientist
- Evaluate wildfire risk of various locations around the world
- Develop and maintain the data pipeline using in the wildfire model
- Develop and maintain the data pipeline using in the wildfire model
Saegus
Jan 2020 - Jun 2021 (1 year 5 months)
Machine Learning Engineer
Near Duplicate Video Detection (L'Oréal)
- Find twice many relevant results with a modification in the main algorithm.
- Reduce processing time 5 times using cache and parallel computing.
- Scale the system to be able to search from ten of thousands of videos.
One Shot Learning with Siamese Networks (BIC)
- Reduce model training time ~ 10 - 15 times using various optimization and parallel processing techniques.
- Increase model accuracy from 50% to 90%.
- Handle high loads with asynchronous API.
- Automate development and deployment process using DevOps best practices.
- Find twice many relevant results with a modification in the main algorithm.
- Reduce processing time 5 times using cache and parallel computing.
- Scale the system to be able to search from ten of thousands of videos.
One Shot Learning with Siamese Networks (BIC)
- Reduce model training time ~ 10 - 15 times using various optimization and parallel processing techniques.
- Increase model accuracy from 50% to 90%.
- Handle high loads with asynchronous API.
- Automate development and deployment process using DevOps best practices.
Heavenize
Apr 2019 - Sep 2019 (5 months)
Data Scientist
Subject: “R&D Artificial Intelligence and Market Finance.”
- Design a Data Science pipeline to preprocess, train and evaluate Machine Learning models for financial data.
- Apply Machine Learning algorithms to automate a market prediction process.
- Survey the state-of-the-art of market prediction using news and sentiment analysis.
- Design a Data Science pipeline to preprocess, train and evaluate Machine Learning models for financial data.
- Apply Machine Learning algorithms to automate a market prediction process.
- Survey the state-of-the-art of market prediction using news and sentiment analysis.
Add some compelling projects here to demonstrate your experience
Quantlib: Enable time-dependent steps for binomial tree engines
Dec 2018 - Feb 2019
- Annalyse the bottleneck for the current binomial tree class with time-dependent parameters.
- Boost results to 3-4 times faster by implementing a cache solution and a vector solution to cache common results.
======================
"The QuantLib project (http://quantlib.org) is aimed at providing a comprehensive software framework for quantitative finance. QuantLib is a free/open-source library for modeling, trading, and risk management in real-life."
- Boost results to 3-4 times faster by implementing a cache solution and a vector solution to cache common results.
======================
"The QuantLib project (http://quantlib.org) is aimed at providing a comprehensive software framework for quantitative finance. QuantLib is a free/open-source library for modeling, trading, and risk management in real-life."
Deep Learning with Long Short-Term Memory Networks for Financial Market Prediction
Nov 2018 - Mar 2019
- Reproduce the result of the paper "Deep Learning with Long Short-Term Memory Networks for Financial Market Prediction"
- Scrape, clean and explore a new dataset.
- Create an LSTM model that is better than the benchmark in the same trading period.
- Discuss the architecture limitation of the benchmark architecture and propose solutions.
- Scrape, clean and explore a new dataset.
- Create an LSTM model that is better than the benchmark in the same trading period.
- Discuss the architecture limitation of the benchmark architecture and propose solutions.
Dematerialization a local complementary currency
Feb 2017 - Jun 2017
In this project, we research and provide technical support to an association of local complementary currency. The aim of this project is to provide a secure online tools for transactions.
This section lets you add any degrees or diplomas you have earned.
IMT Atlantique
Master of Engineering - MEng, Data Science for Finance
Dec 2015 - Dec 2018
Subject: Signal Processing, Computer Science, Machine Learning, Quantitative Finance.
Ho Chi Minh University of Technology
Bachelor's degree, Telecommunications Engineering
Dec 2011 - Dec 2015
Excellent Engineering Training Program (Programme de Formation d'Ingénieurs d'Excellence au Vietnam)
Mathematics for Machine Learning Specialization (Imperial College London)
Apr 2020
Advanced Linear Models for Data Science 1: Least Squares (Johns Hopkins University)
Jul 2020
Advanced Machine Learning with TensorFlow on Google Cloud Platform Specialization (Google Cloud)
Aug 2020