Leonardo Leon Vera, Developer in Lima, Callao Region, Peru
Leonardo is available for hire
Hire Leonardo

Leonardo Leon Vera

Verified Expert  in Engineering

Artificial Intelligence Computer Scientist and Developer

Lima, Callao Region, Peru

Toptal member since February 22, 2024

Bio

Leonardo is a computer scientist specializing in machine learning technologies. His expertise lies in developing cutting-edge image recognition applications for embedded and on-premise platforms. Passionate for forefront technological solutions, Leonardo embodies professionalism with his responsibility and self-directed learning. As a proactive individual, he is continually motivated to challenge himself by establishing result-oriented objectives that push the boundaries of his capabilities.

Portfolio

Atos
Computer Vision, Big Data, DeepStream SDK, NVIDIA TAO, OpenCV, NVIDIA TensorRT...
GETTER - Amplified Industry
Computer Vision, Deep Learning, Keras, PyTorch, NVIDIA TensorRT, Jetson TX2...
Centro de Tecnologías de la Información y Comunicaciones CTIC UNI
TensorFlow, OpenCV, You Only Look Once (YOLO), Jetson TX2...

Experience

  • Python 3 - 6 years
  • Ubuntu - 6 years
  • Machine Learning - 6 years
  • TensorFlow - 4 years
  • Computer Vision - 4 years
  • Docker - 3 years
  • DeepStream SDK - 2 years
  • NVIDIA TAO - 2 years

Availability

Part-time

Preferred Environment

Ubuntu, Python 3, Vim Text Editor, Visual Studio Code (VS Code), Conda, Docker, Artificial Intelligence (AI), Python

The most amazing...

...thing I've deployed is fatigue detection on a truck using TensorRT and the Jetson platform.

Work Experience

Computer Vision Engineer

2021 - PRESENT
Atos
  • Designed and implemented an automatic MLOps pipeline using Airflow, MLflow, Data Version Control (DVC), and NVIDIA TAO to retrain object detection models in computer vision.
  • Employed OpenCV and traditional computer vision techniques to solve straightforward problems that do not require intensive computations like convolutional neural networks (CNNs).
  • Managed Kubernetes workflows and crafted complete Docker Compose pipelines for computer vision machine learning initiatives.
  • Deployed to production with Triton Inference Server and integrated DeepStream into the abovementioned pipelines.
Technologies: Computer Vision, Big Data, DeepStream SDK, NVIDIA TAO, OpenCV, NVIDIA TensorRT, MLflow, Convolutional Neural Networks (CNNs), Docker Compose, Kubernetes, Artificial Intelligence (AI), Python, Machine Learning Operations (MLOps), Open Source, Databases, Google Sheets, Optical Character Recognition (OCR), Data Pipelines, Neural Networks, Image Processing

Computer Vision Engineer

2019 - 2021
GETTER - Amplified Industry
  • Headed and created machine learning and computer vision projects for education, retail, and ergonomics 4.0.
  • Implemented object detection, image classification, and pose estimation models using Keras and PyTorch.
  • Executed a university project to monitor student attention using face and emotion recognition technologies.
  • Introduced pose estimation technology in the industrial sector to evaluate ergonomic postures.
  • Deployed fatigue detection on a truck using TensorRT and the Jetson platform.
Technologies: Computer Vision, Deep Learning, Keras, PyTorch, NVIDIA TensorRT, Jetson TX2, NVIDIA Jetson, Artificial Intelligence (AI), Python, Open Source, Optical Character Recognition (OCR), Neural Networks, Image Processing

Researcher

2015 - 2019
Centro de Tecnologías de la Información y Comunicaciones CTIC UNI
  • Designed and built an autonomous robot for a robotics competition in Brazil.
  • Authored and presented papers and posters at international conferences.
  • Developed a beginner's deep learning course and researched you only look once (YOLO) implementations for Jetson and Raspberry Pi.
Technologies: TensorFlow, OpenCV, You Only Look Once (YOLO), Jetson TX2, Artificial Intelligence (AI), Python, Open Source, Optical Character Recognition (OCR), Neural Networks

Junior Researcher

2017 - 2017
Artificial Intelligence Group of PUCP
  • Implemented variational autoencoders for reconstructing 3D archaeological objects in a project.
  • Utilized TensorFlow and Keras frameworks to build custom architectures.
  • Contrasted models with generative adversarial networks (GANs) for a reconstruction task.
Technologies: Convolutional Neural Networks (CNNs), OpenCV, Keras, TensorFlow, Artificial Intelligence (AI), Python, Open Source, Neural Networks

Experience

AI-powered Mini Robot for Vehicle Surveillance in Underground Parking Facilities

Spearheaded the construction of a compact autonomous vehicle with NVIDIA Jetson TX1, Arduino, and cameras, employing deep learning for self-navigation, vehicle detection, and license plate recognition in underground garages, complemented by beacon sensors for parking lot localization.

Education

2021 - 2024

Master's Degree in Computer Science

Pontifical Catholic University of Parana - Parana, Brazil

2013 - 2019

Bachelor's Degree in Computer Science

National University of Engineering - Lima, Peru

Certifications

OCTOBER 2019 - PRESENT

English EF SET Certificate

EF SET

Skills

Libraries/APIs

OpenCV, TensorFlow, PyTorch, Keras

Tools

Vim Text Editor, NVIDIA TAO, Apache Airflow, Git, You Only Look Once (YOLO), Jetson TX2, Docker Compose, NVIDIA Jetson, Google Sheets

Languages

Python 3, Bash, Python, C++

Platforms

Ubuntu, Docker, Visual Studio Code (VS Code), Kubernetes, Arduino

Storage

Databases, Data Pipelines

Other

Machine Learning, Computer Vision, Conda, DeepStream SDK, Artificial Intelligence (AI), Open Source, K-nearest Neighbors (KNN), Optical Character Recognition (OCR), Neural Networks, Image Processing, MLflow, Deep Learning, Natural Language Processing (NLP), Convolutional Neural Networks (CNNs), Big Data, NVIDIA TensorRT, English, Robotics, Jetson TX1, Machine Learning Operations (MLOps)

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