Ondřej is available for hire
Hire OndřejOndřej Pacovský
Verified Expert in Engineering
Machine Learning Developer
Location
Prague, Czech Republic
Toptal Member Since
May 21, 2018
Ondřej enjoys pushing the boundaries of artificial intelligence and making it useful for humanity. Ondřej has wide-ranging experience in artificial intelligence and machine learning, from being an individual contributor to leading teams and launching his own companies.
Portfolio
RailState
Machine Learning, Computer Vision, Python, Management, Team Leadership...
Eyen SE
Deep Learning, Machine Learning, Scanning Electron Microscopes, TensorFlow...
Diffbot
Computer Vision, Kotlin, Storage, Machine Learning, Artificial Intelligence (AI)
Experience
Availability
Part-time
Preferred Environment
C++, Git, Linux, Pandas, NumPy, Python, TensorFlow, Kotlin
The most amazing...
...thing I've participated in is the Gmail Priority Inbox, saving time sifting through email for millions of users.
Work Experience
Chief Technology Officer, Co-founder
2021 - PRESENT
RailState
- Co-founded the RailState project, bringing in software and AI expertise.
- Delivered MVP timely, which led to first clients and Series A funding.
- Led the software team implementing a full version of the IoT pipeline, now running 100+ devices in North America.
- Collaborated with the sales team, customers, and domain experts to come up with the most important features to focus on.
- Coordinated hardware and ops teams to expand our network and maintain the quality of service.
Technologies: Machine Learning, Computer Vision, Python, Management, Team Leadership, Artificial Intelligence (AI), Jetson TX2
Co-founder, CEO, Chief Data Scientist
2013 - PRESENT
Eyen SE
- Raised capital and put together an exceptional development team.
- Developed a deep neural network-based system for automatically processing particles in micrographs from an electron microscope.
- Implemented a correlation-based alignment method on GPUs, 100x faster than original NumPy/SciPy code.
Technologies: Deep Learning, Machine Learning, Scanning Electron Microscopes, TensorFlow, NumPy, Python, NVIDIA CUDA, C++
Senior Machine Learning Engineer
2020 - 2022
Diffbot
- Developed a system for cost-efficient storing, analyzing, and serving hundreds of millions of images on custom-built dedicated hardware.
- Designed, trained, and deployed a system for rating the quality of images.
- Optimized throughput of the computer vision inference engine using NVIDIA Triton Inference Server and load balancer to optimize cost per scored image.
Technologies: Computer Vision, Kotlin, Storage, Machine Learning, Artificial Intelligence (AI)
Software Engineer
2009 - 2013
- Brought Priority Inbox from prototype to public launch.
- Redesigned and prototyped a new learning back-end.
- Managed Bigtable replication and load balancing across the globe.
Technologies: Semantic Analysis, Machine Learning, MapReduce, BigTable, Java, Python, C++
Software Engineering Intern
2008 - 2008
- Reduced the estimation error by 40% compared to handcrafted models.
- Sped up the data acquisition library 10x for specific query types.
Technologies: MapReduce, C++, Machine Learning
Lead Developer, AI Specialist
2007 - 2007
ALTAR Interactive
- Led eight developers to deliver a successful, timely prototype.
- Designed and implemented an AI module for the prototype.
- Grew the team and introduced systematic development techniques.
Technologies: SWIG, Python, C++
Experience
GMail Priority Inbox
https://www.google.com/mail/help/priority-inbox.htmlPriority Inbox automatically identifies your important email and separates it from everything else so you can focus on what really matters. I implemented a part of the learning back end and designed a second, more efficient generation of the distributed scoring algorithm.
Eyen - Celine 3D Cryo-electron Microscopy Service
http://www.eyen.euEyen is the expert in molecular imaging and nano-scale structural data analysis. Our expertise lies in advanced computational analysis of proteins, biologics, and nanoparticles for pharmaceutical research and development.
Our unique 3D cryo-electron microscopy service supported by advanced algorithms provides outsourcing for your protein, biologics, and nanoparticle structural characterization needs quickly and with low up-front costs.
Our unique 3D cryo-electron microscopy service supported by advanced algorithms provides outsourcing for your protein, biologics, and nanoparticle structural characterization needs quickly and with low up-front costs.
RailState
http://www.railstate.comA system of edge IoT devices monitoring rail traffic across North America. The specialized cameras capture information throughout the day and night, infer important information using GPU-accelerated computer vision algorithms, and upload the information to the cloud. We then analyze and model the data and present it via a web application.
Skills
Languages
Python, C++, Java, Prolog, Kotlin
Libraries/APIs
NumPy, TensorFlow, OpenCV, Pandas
Platforms
Linux, NVIDIA CUDA
Storage
BigTable, Google Bigtable
Other
Deep Learning, Natural Language Processing (NLP), Word2Vec, Machine Learning, Deep Neural Networks, Neural Networks, GPT, Generative Pre-trained Transformers (GPT), Computer Vision, Scanning Electron Microscopes, Semantic Analysis, Robotics, Algorithms, Evolutionary Computation, Artificial Intelligence (AI), Storage, Team Leadership, Convolutional Neural Networks (CNN)
Tools
Git, SWIG, Jetson TX2
Paradigms
GPGPU, MapReduce, Management
Education
2000 - 2006
Master's Degree in Theoretical Computer Science
Charles University - Prague, Czech Republic
2004 - 2005
Master of Science Degree (Distinction) in Evolutionary and Adaptive Systems
Sussex University - Brighton, UK
Certifications
NOVEMBER 2018 - PRESENT
Deep Learning Specialization
Coursera
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