Jurgis Pasukonis, Developer in Vilnius, Vilnius County, Lithuania
Jurgis is available for hire
Hire Jurgis

Jurgis Pasukonis

Verified Expert  in Engineering

Data Engineer and Developer

Location
Vilnius, Vilnius County, Lithuania
Toptal Member Since
May 20, 2020

Along with holding a Ph.D. in theoretical physics and an undergraduate degree from MIT, Jurgis is an expert in fields ranging from AI and deep learning to data warehousing and back-end infrastructure. With 15 years of experience in the industry, he's occupied multiple roles, including acting as the CTO and co-founder of two companies.

Data EngineeringMachine LearningArtificial Intelligence (AI)C#SQLAmazon Web Services (AWS)PythonMicroservicesTableauDeep Neural NetworksBusiness Intelligence (BI)OLAPTensorFlowPyTorchBig DataVertica

Portfolio

Three Thirds
Kubernetes, PyTorch, Machine Learning, Deep Neural Networks, Deep Learning...
Trafi
Amazon Web Services (AWS), TensorFlow, Google BigQuery, Docker, iOS, Android...
Adform
Amazon Web Services (AWS), Vertica, Hadoop, Python, Machine Learning...

Experience

C# - 15 yearsData Engineering - 10 yearsAmazon Web Services (AWS) - 7 yearsMachine Learning - 5 yearsPython - 5 yearsData Science - 5 yearsArtificial Intelligence (AI) - 2 yearsDeep Reinforcement Learning - 2 years

Availability

Part-time

Preferred Environment

SQL, C#, Python, TensorFlow, Amazon Web Services (AWS)

The most amazing...

...thing I've created is a mobility platform company called Trafi, from initially developing its core algorithms to hiring and running a world-class team.

Work Experience

Co-founder | Lead AI Researcher

2019 - 2021
Three Thirds
  • Developed a novel model-based reinforcement learning agent, which learns to predict its environment.
  • Set up a scalable infrastructure to run training jobs on GPU using Google Cloud, a Kubernetes cluster, and an MLflow tracking server.
  • Created a machine-learning adoption strategy for a big enterprise client.
  • Created proof-of-concept computer vision solution for a client, using convolutional deep networks.
Technologies: Kubernetes, PyTorch, Machine Learning, Deep Neural Networks, Deep Learning, Artificial Intelligence (AI), Computer Vision

Co-founder | CTO

2013 - 2020
Trafi
  • Co-founded the company, which develops mobile apps for public transport used by over 1 million users.
  • Hired and led a team of 30 engineers that developed Android, iOS, and web apps (all powered by a sophisticated back-end platform).
  • Developed a state-of-the-art real-time multi-modal public transport routing algorithm. It relied on high-performance graph search implementation, augmented with statistical modeling.
  • Built a deep learning model to predict which of the possible public transport routes would be preferred by a user, based on historical usage patterns.
  • Managed the AWS IT infrastructure to enable fast development and scalable performance. This included architecture and set up of container clusters, networking, databases, queues, caches, CI pipelines, and so on.
  • Developed the whole back-end platform to process the data, perform routing, and serve the app via a REST API, using C#, and deployed as microservices on AWS.
Technologies: Amazon Web Services (AWS), TensorFlow, Google BigQuery, Docker, iOS, Android, C#, Machine Learning, Data Science, Business Intelligence (BI), Big Data, Microservices

Lead Data Scientist

2009 - 2013
Adform
  • Established a big data infrastructure in the company using Hadoop on Amazon EMR (AWS Elastic MapReduce) for the ETL and Vertica for the data warehouse.
  • Developed predictive models of user behavior when interacting with ads, which were used in production to set prices in real-time bidding.
  • Hired and led a team of data scientists, which drove the adoption of machine learning in the company.
Technologies: Amazon Web Services (AWS), Vertica, Hadoop, Python, Machine Learning, Data Science, Business Intelligence (BI), Big Data

Senior Back-end Developer

2006 - 2008
Adform
  • Developed a core back-end infrastructure, which enabled the processing of very high data volumes using MS SQL and C#.
  • Led a project to replace an existing SQL data warehouse with an OLAP-based solution using MS Analysis Services.
  • Developed a reporting engine, which could automatically generate reports based on a dynamic definition using MS Reporting Services.
Technologies: OLAP, SQL, C#

Founding and Building Trafi

http://www.trafi.com
A mobility platform, consisting of mobile apps (Android, iOS) for people to navigate with public transport in the city, a data processing backend for various types of real-time mobility data, and a unique multimodal routing algorithm.

As a founder and CTO of the company, I oversaw the development of all parts of the system, designed the architecture, hired and led teams, and personally developed the routing algorithm and other key parts of the back-end infrastructure.

Big Data Infrastructure for Adform

I was solely responsible for a pilot big data project, which processed terabytes of data on a Hadoop cluster and trained ML models to predict ad-clicking behavior. It involved creating and managing AWS EMR infrastructure, writing ETL jobs in Hive and Python, training ML models with scikit-learn, and pushing data to Vertica data warehouse.

AI Strategy for Girteka

Girteka is one of the largest logistics companies in Europe. As a contractor, I owned the analysis of their data infrastructure and data quality as well as roadmap creation for the use of machine learning throughout the system. The project culminated in a pilot implementation of one of the ML components.
2009 - 2013

Ph.D. in String Theory

Queen Mary University of London - London, United Kingdom

2008 - 2009

Master's Degree in Theoretical Physics

Imperial College London - London, United Kingdom

2001 - 2005

Bachelor's Degree in Physics

MIT | Massachusetts Institute of Technology - Cambridge, MA, United States

Languages

C#, Python, SQL

Tools

Tableau

Paradigms

Data Science, Microservices, OLAP, Business Intelligence (BI), Requirements Analysis

Platforms

Amazon Web Services (AWS), Docker, Apache Kafka, Android, iOS, Kubernetes

Other

Machine Learning, Google BigQuery, Data Engineering, Performance Optimization, Artificial Intelligence (AI), Big Data, Technical Leadership, Data Analytics, Deep Neural Networks, Deep Reinforcement Learning, Computer Vision, Deep Learning, Classification Algorithms, Regression, Image Recognition

Libraries/APIs

TensorFlow, PyTorch, Scikit-learn

Storage

Vertica

Frameworks

Hadoop

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