Doru Musuroi, Developer in Zürich, Switzerland
Doru is available for hire
Hire Doru

Doru Musuroi

Verified Expert  in Engineering

Software Engineering Developer

Location
Zürich, Switzerland
Toptal Member Since
April 18, 2019

Doru is a highly proficient data scientist with proven experience in Pandas, Spark, Kafka, Seaborn, Plotly, PyTorch and Scikit-learn. He's performed data analysis on large-scale datasets at Oracle, worked on distributed systems at Facebook, and designed and implemented big data platforms at Dathena. Doru has also completed his master’s degree in computer science at a top ten university, the Swiss Federal Institute of Technology Lausanne.

Software EngineeringPythonPandasSQLAgile Software DevelopmentJavaSparkPyTorchKerasApache KafkaHDFSDistributed SystemsAmazon Web Services (AWS)LinuxGit

Availability

Part-time

Preferred Environment

Git, PyTorch, Python, Pandas, Spark, Linux, Docker, Google Cloud Platform (GCP), Amazon Web Services (AWS)

The most amazing...

...thing I've coded is a probabilistic journey planner (like Google Maps) that suggests journeys based on a probabilistic model of train delays in Switzerland.

Work Experience

Machine Learning Research Intern

2019 - PRESENT
Oracle
  • Performed data analyses on large-scale datasets.
  • Implemented a graph analysis with PGX, a parallel graph processing engine.
  • Researched about graph machine learning for semantic embeddings of the graph topology.
Technologies: Python, Dask, Pandas, PGX, Spark

Data Science Intern

2018 - 2019
Dathena
  • *Improved Spark job performance using profiling tools such as Sparklen and pinpointed bottlenecks.
  • Analyzed the bottlenecks of the core big-data platform on top of Spark and redesigned it from a batch-processing architecture to a streaming one.
  • Implemented a stream-based big data platform for file ingestion using Kafka, Spark Streaming, Hadoop, and HBase.
Technologies: Apache Kafka, Spark, HBase, Hadoop, Scala

Software Engineer Intern

2018 - 2018
Facebook
  • Developed a Python service that reduced–by 83%–the latency of requests inside the CI system.
  • Implemented a Python service that improved the processing time of the requests inside the CI system by 90 ms, resulting in the saving of 1,400 hours of computing time per day.
  • Extended a package management tool to enable the automatic deployment of Windows services.
Technologies: Python

Software Development Engineer Intern

2016 - 2016
Amazon
  • Implemented an Eclipse plugin that reduced the writing time for the config files of an internal service by 70% and removed the possibility of creating mistaken config files.
Technologies: Eclipse, Java

Robust Journey Planner

Here, I implemented a robust journey planner (think Google Maps) that recommends journeys along with a likelihood percentage of reaching the destination. I used historical data from the Swiss national transportation system and created a probabilistic model that would predict the probability of delays for a train at each station. I then used this model on top of OpenTripPlanner in order to obtain journey recommendations that would respect the desired certainty of reaching the destination.

Violent Scene Detection in Movies

This is a research project involving computer vision working on the task of detecting sequences of frames containing physical violence. I designed and implemented a deep learning pipeline (in PyTorch) to explore the limits of temporal segment networks on various violent scene detection benchmarks.
2017 - 2019

Master's Degree in Computer Science

Swiss Federal Institute of Technology - Zürich, Switzerland

2014 - 2017

Bachelor's Degree in Computer Science

University of Bucharest - Bucharest, Romania

Languages

Python, Java, SQL, Scala, JavaScript

Libraries/APIs

Pandas, Dask, PyTorch, Keras, Spark ML, Spark Streaming

Platforms

Jupyter Notebook, Apache Kafka, Zeppelin, Linux, Eclipse, Docker, Google Cloud Platform (GCP), Amazon Web Services (AWS)

Other

Software Engineering, PGX, Distributed Systems

Frameworks

Spark, Hadoop, Spark Structured Streaming

Paradigms

Agile Software Development, Concurrent Programming, Functional Programming, Data Science

Storage

HDFS, HBase

Tools

Git, Spark SQL

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring