About Me
As a data enthusiast and Python developer, I have been involved in various data-related projects for almost a decade. With an M.Sc. in Computer Science focused on Data Science and a Bachelor's degree in Computer Science, I have expertise in areas such as recommender systems, natural language processing, data stream mining, and computer vision. Some of the tecnologies that I've already worked on are PyTorch, TensorFlow, AutoML, Sparkling Water, spaCy, NLTK, and etc.
Data Scientist & ML Engineer
- Birthday: 21 Aug 1997
- Website: heise.sh
- City: Curitiba, Paraná, Brazil
- Age: 26
- Degree: Master of Science
- E-mail: eduheise@gmail.com
- Freelance: Contact for Availability
I am passionate about my work and strive to deliver impactful solutions using data science and machine learning. With a strong educational background and experience, I am well-equipped to tackle complex challenges.
Skills Overview
Resume
Sumary
Eduardo Heise
Accomplished and results-oriented Data Scientist with a proven track record of delivering innovative and data-driven solutions. Possessing almost a decade of experience in designing and implementing advanced predictive models, I excel in transforming complex data into actionable insights.
- Curitiba, Paraná, BR
- eduheise@gmail.com
Education
Master of Science in Applied Informatics
2018 - 2020
Pontifical Catholic University of Paraná (PUCPR), Curitiba, PR
Focused on Data Science and Recommender Systems. GPA: 3.86 out of 4.0.
Bachelor of Computer Science
2016 - 2020
Pontifical Catholic University of Paraná (PUCPR), Curitiba, PR
Awarded the prestigious prize for Best Academic Performance with a GPA of 3.32 out of 4.0.
Licenses and Certificates
Natural Language Processing
Generative AI with Large Language Models
DeepLearning.ai
Deep Learning
Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization
DeepLearning.ai
Data Science and Health Sciences
Summer School on Data Science in Health and Disease
Leiden University Medical Center
Professional Experience
LLM Engineer
Oct. 2023 — Present
Andela Inc., New York, NY
- Developed a RAG-based application using LangGraph, LangChain, and VertexAI to assess talent qualifications and transform client requests into actionable job requirements.
- Trained a PEFT (LoRA) adapter to fine-tune large language models, significantly enhancing performance in staffing applications.
- Applied OpenAI Whisper for speech-to-text transcription, processing recordings with LLMs to extract valuable insights.
- Worked with diverse teams to identify key challenges, source relevant data, propose innovative LLM-based solutions, and present user-friendly interfaces focused on usability and functionality.
Deep Learning Engineer
Oct. 2022 — Oct. 2023
Index Group, Remote
- Developed advanced semantic segmentation models, reducing a 4-day manual task to 30 minutes, significantly boosting efficiency and productivity.
- Processed high-resolution images (>5B pixels per image) using scalable infrastructure for computer vision tasks, achieving high-performance and cost-effective predictions.
- Improved prediction accuracy from 85% to 95% by applying post-processing techniques, such as morphological operations, to eliminate data noise.
- Built a dynamic interface for real-time model interaction, enabling users to visualize and engage with the geospatial product effectively.
Lead Data Scientist
Sep. 2021 — Oct. 2022
Sumersoft Technology, Remote
- Led a team of developers and data scientists, actively coding to design and launch the first prototype of a cutting-edge IoT intelligent product.
- Deployed intelligent systems on devices with 520KB RAM, optimizing architecture and using lightweight algorithms for efficient performance.
- Conducted experiments with Histogram of Oriented Gradients (HOG) to improve feature extraction, enhancing model accuracy for IoT applications.
- Assessed and implemented TinyML solutions with embedded TensorFlow, enabling efficient machine learning model deployment on low-power microcontrollers.
Senior Data Scientist
Sep. 2020 — Sep. 2021
Melo Advocacy, Remote
- Utilized BERT-based semantic vectors to train models for accurate text classification, significantly enhancing textual data processing capabilities.
- Developed solutions using large language models (LLMs) to create tailored juridical documents based on stakeholder requirements.
- Designed a self-training NLP system applied to thousands of legal documents, automating lead identification and driving client base expansion.
Data Scientist
Apr. 2016 — Sep. 2020
Other Experiences, Remote
- Supported NLP projects by applying bag of words, TF-IDF, and Word2Vec techniques to process textual data, contributing to improved model accuracy under senior guidance.
- Assisted in handling high-imbalanced datasets for sentiment analysis, using traditional ML methods and over/undersampling techniques to achieve reliable results.
- Worked on NLP tasks in the publishing industry, aiding in predicting exam question subjects for a classification task involving over 1,000 classes.
- Contributed to the development of online recommender systems, helping implement adaptive algorithms to address concept drift with team support.
- Leveraged PySpark to handle large datasets with millions of records, successfully developing AI-based products tailored to client needs.
