Diego Kiedanski, Developer in Montevideo, Montevideo Department, Uruguay
Diego is available for hire
Hire Diego

Diego Kiedanski

Verified Expert  in Engineering

Software Developer

Montevideo, Montevideo Department, Uruguay
Toptal Member Since
February 8, 2023

Diego is a Computer Engineer specializing in internet measurements and data analytics for internet traffic patterns. He has worked as a machine learning consultant on projects ranging from price optimization for insurance companies to remote sensing and computer vision to detect schools in remote areas for UNICEF. Diego has worked on three continents, from Yale University (USA) to national train companies (Luxembourg) and South America.


Python 3, SQL, React, JavaScript, PyTorch, Scikit-learn, Machine Learning...
Yale University
PyTorch, Python 3, Blockchain, Image Processing, Computer Vision, Deep Learning...
Optimization, Convex Optimization, Gurobi, CPLEX, Reinforcement Learning...




Preferred Environment

Slack, Python 3, Visual Studio Code (VS Code), PyTorch, Bash Script, Scikit-learn

The most amazing...

...product I've developed consists on a machine learning model that improves the driving experience by improving automatic gear shifting.

Work Experience

Lead Machine Learning Engineer

2021 - 2023
  • Created a new standard for data management pipelines that was quickly incorporated into many of the company's projects. These practices reduced 30% of the time spent on data management.
  • Collaborated with significant consulting companies to implement and deliver their AI strategy. These allowed me to work on large-scale projects involving more than 30 stakeholders. I was the technical lead for the machine learning components.
  • Designed and developed highly efficient machine learning models that improve drivability by reading real-time measurements from a car in motion. The deliverable also included a dashboard that was automatically updated every few milliseconds.
Technologies: Python 3, SQL, React, JavaScript, PyTorch, Scikit-learn, Machine Learning, Image Processing, Computer Vision, Stable Diffusion, Deep Learning, Python, APIs, Google Cloud Platform (GCP), Amazon Web Services (AWS), Artificial Intelligence (AI), Data Science, Text Processing, Text Generation, Early-stage Startups, Google Cloud, Optimization, Data Analytics, Full-stack Development, ARIMA, ARIMA Models, LSTM, SARIMA, Supply Chain Management (SCM), Supply Chain Optimization, Forecasting, Web Development, Internet of Things (IoT), Large Language Models (LLMs)

Postdoctoral Associate

2022 - 2022
Yale University
  • Designed a framework to apply machine learning of Decentralized Finance transactions using neural graph networks. Decentralized finance is a financial protocol that runs on the Ethereum blockchain.
  • Used federated learning to train distributed machine learning models with applications to air quality measurements.
  • Implemented benchmarks for different computer vision data loaders, an integral part of the training pipeline in deep learning projects.
Technologies: PyTorch, Python 3, Blockchain, Image Processing, Computer Vision, Deep Learning, Python, APIs, Amazon Web Services (AWS), Machine Learning, Artificial Intelligence (AI), Data Science, Data Analytics, Full-stack Development, Forecasting

Optimisation Engineer

2020 - 2020
  • Acted as the lead researcher on a European project involving state-of-the-art local energy markets and how new regulations would affect innovation in the sector. I delivered all the expected deliverables using 25% of the expected resources.
  • Represented the company on a joint project to test optimal charging of a fleet of electric vehicles to deliver demand response as a service.
  • Collaborated on a project using reinforcement learning to optimize the use of an energy battery participating in several demand response services at the same time.
Technologies: Optimization, Convex Optimization, Gurobi, CPLEX, Reinforcement Learning, Electric Vehicles, Forecasting

Research Engineer

2018 - 2018
Télécom Paris
  • Developed a simulation environment for agents buying and selling renewable energy in a double-sided auction.
  • Built a prototype using reinforcement learning for bidding in an auction while optimizing under certain constraints, like meeting the demand for energy in a household.
  • Wrote three technical papers at international conferences.
Technologies: Python, Game Theory, Blockchain, Online Auctions, Reinforcement Learning, Deep Reinforcement Learning, Optimization, Data Analytics, Full-stack Development, Forecasting

Full-Stack Developer

2016 - 2016
Universidad de la República del Uruguay
  • Acted as the leading developer of an information system designed to simplify the work of the administrative staff behind the computer engineering degree.
  • Oversaw the migration between a legacy database system and its newer version, both of which we had to maintain to provide the required functionality to the system users.
  • Designed large-scale data transformations involving all records of all courses taken by all students in the university, which could take days to process.
Technologies: Python, SQL, Databases, PostgreSQL, Full-stack Development, Web Development

Parking Lot Occupancy Forecasting Model

Fear of finding a parking lot full had been identified as a critical bottleneck in the uptake of trains over cars for daily commutes. I developed a machine learning model to forecast the parking occupancy at train stations in Luxembourg. Part of the challenge of the project was integrating the model's prediction into the company's existing data platform and workflows.

Tutorial on Object Detection

As part of my previous company's outreach program, I wrote a blog post detailing how to get started in data science and machine learning. In particular, the third blog post of the series contains a guided example of how to set up a computer vision project around object detection.

Lion Identification

The project aimed to develop a computer vision-based system to detect and identify individual lions.
In the same way, humans can be identified by their fingerprints, and the pattern of their whiskers can detect lions.
I was the liaison between Tryolabs and two undergrad students that contributed to the project. In that role, I performed weekly follow-up sessions and overall mentoring.

Vegan-Friendly Detector App

Large language models like GPT-3 and ChatGPT are well known to perform worse in less represented and popular communities.
I used a new platform called brancher.ai to quickly build an app using GPT-3 that tries to infer whether a vegan would agree with a given phrase.
2019 - 2020

Doctorate Degree in Informatics and Applied Mathematics

Polytechnic Institute of Paris - Paris, France

2013 - 2017

Bachelor's Degree in Computer Engineering

University of the Republic - Montevideo, Uruguay


PyTorch, LSTM, Scikit-learn, React


Slack, Gurobi, CPLEX


Python 3, SQL, Python, Bash Script, JavaScript, R




Data Science, Agile Software Development


Google Cloud, Databases, PostgreSQL


Amazon Web Services (AWS), Visual Studio Code (VS Code), Blockchain, Pentaho, Google Cloud Platform (GCP)


Programming, Machine Learning, IT Consulting, Data Analytics, Artificial Intelligence (AI), Computer Vision, Deep Learning, APIs, ARIMA, ARIMA Models, SARIMA, Forecasting, Natural Language Processing (NLP), GPT, Generative Pre-trained Transformers (GPT), Game Theory, Full-stack Development, Technology Consulting, Image Processing, Text Processing, Early-stage Startups, Optimization, Web Development, Internet of Things (IoT), Management Consulting, Business Consulting, Stable Diffusion, Text Generation, Online Auctions, Reinforcement Learning, Deep Reinforcement Learning, Convex Optimization, Electric Vehicles, Supply Chain Management (SCM), Supply Chain Optimization, Large Language Models (LLMs)

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.


Share your needs

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

Choose your talent

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

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