Luis is available for hire
Hire LuisLuis Antunes
Verified Expert in Engineering
Machine Learning Engineer and Developer
Location
Mississauga, Canada
Toptal Member Since
September 27, 2022
Luis is a seasoned engineer with 15 years of software development experience and a strong background in machine learning. As an expert in scientific computing and application development, he has focused on machine learning and data science projects for the past five years. Luis enjoys applying his fundamental research skills and knowledge of machine learning and engineering to solve challenging problems that require predictive analytics.
Portfolio
University of Reading
Python, Keras, TensorFlow, Amazon Web Services (AWS), Flask, NumPy, Matplotlib...
Uncharted Software
Python, Node.js, NumPy, Docker, SciPy, Keras
Deep Genomics
Python, Go, Java, Docker, Kubernetes, HTML, JavaScript, Google Cloud
Experience
Availability
Part-time
Preferred Environment
MacBook, PyCharm, Git, Slack, Amazon Web Services (AWS)
The most amazing...
...project I've worked on is discovering a new approach for learning machine representations of atoms using an idea from natural language processing.
Work Experience
Part-time PhD Candidate
2020 - PRESENT
University of Reading
- Carried out research in machine learning applied to problems in chemistry, with the goal of searching chemical space for new materials for energy applications.
- Developed models of chemical properties with deep neural networks, as well as a novel approach for building distributed representations of atoms and compounds, named SkipAtom, based on the skip-gram model of NLP.
- Introduced a new approach for computational screening of millions of materials using attention-based neural networks, leading to the discovery of many potential new compounds for energy applications, which generally require decades to develop.
Technologies: Python, Keras, TensorFlow, Amazon Web Services (AWS), Flask, NumPy, Matplotlib, JavaScript
Senior Machine Learning Engineer
2020 - 2021
Uncharted Software
- Provided machine learning engineering and research expertise on several projects, including an effort to cluster, summarize, and surface emergent narratives on Twitter and to identify Twitter bot accounts.
- Carried out research in the unsupervised learning of distributed graph representations to facilitate the visualization of large-scale knowledge graphs.
- Researched and implemented stop detection and anomalous trip detection algorithms for GPS datasets using state-of-the-art clustering approaches published in the literature.
- Implemented a system that detects prominent cell phone call networks in CDR datasets using association analysis with the Apriori algorithm.
Technologies: Python, Node.js, NumPy, Docker, SciPy, Keras
Senior Software Engineer
2019 - 2020
Deep Genomics
- Supported the work of biologists by contributing to and maintaining a microservices architecture for storing and managing oligonucleotide data and experimental results, as well as developing a web application interface for querying the data.
- Devised the notation used internally for representing oligonucleotide chemical modifications, which required expert chemical knowledge, and allowed for the proper communication and development of thousands of oligonucleotides.
- Introduced the use of Monte Carlo Tree Search for searching oligonucleotide sequence space for sequences with lower toxicity.
- Participated in developing internal tools for training deep learning models in the cloud.
Technologies: Python, Go, Java, Docker, Kubernetes, HTML, JavaScript, Google Cloud
Senior Software Engineer
2017 - 2019
Thomson Reuters
- Led development efforts, supported researchers, and conducted research in machine learning and reinforcement learning as a member of the Center for AI and Cognitive Computing.
- Served as the lead developer for a major recommendation product on a team of eight software engineers and four research scientists.
- Acted as the lead developer for an anomalous text detection tool, used to highlight interesting SEC filings, on a team consisting of a software engineer and a research scientist.
- Carried out research in neural sentence tokenization, which used recurrent neural networks and word embeddings, and developed a model for an in-house, domain-specific sentence tokenizer.
Technologies: Java, Python, Amazon Web Services (AWS), TensorFlow, Elasticsearch, Hadoop, GPT, Natural Language Processing (NLP), Generative Pre-trained Transformers (GPT), Natural Language Toolkit (NLTK)
Senior Software Engineer
2011 - 2017
Guidewire
- Acted as the technical lead and principal developer for the Guidewire Mississauga office's automated transformation tooling project from March 2016 to April 2017.
- Designed and led the implementation of a deep feed-forward neural net-based system for automatically classifying third-party insurance forms.
- Served as the technical lead of two teams of 15 engineers, spread between Guidewire's Mississauga and Krakow, Poland offices.
Technologies: Java, Gosu, JavaScript, Docker, Amazon Web Services (AWS), Node.js, Amazon S3 (AWS S3)
Experience
SkipAtom
https://github.com/lantunes/skipatomSkipAtom is a software tool I created that scientists can use to learn useful machine representations of atoms. It is inspired by the Word2vec algorithm from NLP. In addition to conceiving the approach and the project, I wrote the TensorFlow and Keras code for learning the representations, wrote scripts that can be used to develop the representations using multiple cores, and published the tool as a package on PyPI. This work was published in a prestigious scientific journal.
CellPyLib
https://github.com/lantunes/cellpylibCellPyLib is a software library I developed for working with cellular automata in Python. Cellular automata are computational systems that are mostly studied in academia but have also found practical use in commercial and industrial settings. I found that existing Python libraries were lacking in terms of robustness, utility, and flexibility, so I decided to develop a new solution. The work was published in the Journal of Open Source Software, and the GitHub project has received over 100 stars. It has been featured in books on computing and used in academic projects. I conceived the project, wrote all the code, and produced all the documentation.
Skills
Languages
Python, Java, SQL, Go, HTML, JavaScript
Libraries/APIs
TensorFlow, Scikit-learn, PyTorch, Keras, NumPy, Node.js, SciPy, Natural Language Toolkit (NLTK), Matplotlib
Tools
PyCharm, Git, Slack
Paradigms
Data Science
Platforms
Amazon Web Services (AWS), Docker, Kubernetes
Storage
NoSQL, Google Cloud, Elasticsearch, Amazon S3 (AWS S3)
Other
Computer Science, Machine Learning, Software Engineering, Technical Writing, Communication, MacBook, Artificial Intelligence (AI), Chemistry, Materials Science, Algorithms, Data Structures, Natural Language Processing (NLP), Gosu, GPT, Generative Pre-trained Transformers (GPT)
Frameworks
Flask, Hadoop
Education
2002 - 2004
Master of Science Degree in Chemistry
York University - Toronto, Canada
1998 - 2002
Bachelor of Science (Hons) Degree in Chemistry
University of Toronto - Toronto, Canada
Certifications
SEPTEMBER 2007 - PRESENT
Technical Communication Graduate Certificate
Seneca College
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