Silas Yunghwa Liu

About

Machine Learning, LLMs and Knowledge Graphs

Data Scientist specialist in LLMs and AI systems, with experience delivering Machine Learning, Data Science and Visualization solutions for insurance, public services, energy and retail. Also Professor and Advisor of a Data Science MBA.

Proficient in Python, Cypher and SQL, with expertise in Knowledge Graphs, Large Language Models, Time Series, Neural Networks and PySpark.

Portuguese · English · German

2016

PhD, Electrical Engineering

UFMG, Brazil — incl. 1-year doctoral stay at TU Dresden, Germany

2011

Master, Electrical Engineering

UFMG, Brazil

2008

Bachelor, Electrical Engineering

UFMG, Brazil

Courses / Certificates

Certificates

May 2025

Hugging Face:
Agents Course

Introduction to Agents
Frameworks for AI Agents
Use Case for Agentic RAG
Final Project - Create, Test and Certify Your Agent

View certificate ↗

August 2024

Udemy:
Graph Theory Algorithms

Graph Theory Introduction and Basics
Graph Theory and Trees
Classic Graph Theory Algorithms
Network Flow
Bonus Topics

View certificate ↗

January 2024

Activeloop / Towards AI / Intel Disruptor Initiative:
Retrieval Augmented Generation for Production with Langchain & LlamaIndex

Basics of Retrieval Augmented Generation
Advanced Retrieval Augmented Generation
Retrieval Augmented Generation and Agents
Retrieval Augmented Generation Evaluation and Observability
Course Extras

View certificate ↗

October 2023

Neo4j GraphAcademy:
Neo4j Graph Data Science Certification

Introduction to Neo4j Graph Data Science library
Neo4j Graph Data Science Fundamentals
Path Finding with GDS
Community Detection with GDS

View certificate ↗

August 2023

edX IsraelX:
Essentials of Genomics and Biomedical Informatics

Biological Sequence Analysis
Biological Databases and How to Search Them
Next Generation Sequencing
Medical Machine Learning
Artificial Intelligence in Medicine
Systems Biology
Human Genomic Variations
Cancer as a Genomic Disease

View certificate ↗

July 2023

Google Skillshop:
Google Analytics Certification

Setup a Google Analytics 4 property for a website or an app
Collect the data for business
Use various reporting tools and features
Recognize key measurement features
Explore insights and make digital marketing decisions

View certificate ↗

June 2023

Neo4j GraphAcademy:
Neo4j Certified Professional

Neo4j Fundamentals
Cypher Fundamentals
Graph Data Modeling Fundamentals
Importing CSV Data into Neo4j
Intermediate Cypher Queries
Building Neo4j Applications with Python

View certificate ↗

February 2023

Datacamp:
Machine Learning Scientist with Python Track (93 H)

Supervised Learning Models
Unsupervised Learning Models
Tree-Based Models
Ensemble Models
Time Series Data
Natural Language Processing
Neural Networks, Deep Learning
Computer Vision

View certificate ↗

December 2022

Datacamp:
Professional Data Scientist

SQL Skills
Data Manipulation
Statistical Fundamentals
Importing & Cleaning
Data Management
Machine Learning Models
Python Programming
Technical Report
Non-Technical Oral Presentation

View certificate ↗

November 2022

Digital Innovation One:
Unimed-BH Data Science

Python
SQL / NoSQL
AWS
Machine Learning

View certificate ↗

September 2022

Datacamp:
Associate Data Scientist

Data Manipulation
Statistical Fundamentals
Importing & Cleaning
Data Management
Machine Learning Fundamentals
Python Programming

View certificate ↗

September 2022

Datacamp:
Data Scientist with Python Track (96 H)

Data Manipulation
Statistical Fundamentals
Importing & Cleaning
Data Visualization
Data Management
Machine Learning Models
Python Programming

View certificate ↗

July 2022

edX MITx 6.86x:
Machine Learning with Python-From Linear Models to Deep Learning (150 H)

Linear Classifier and Perceptron
Generalization
Linear Regression
Nonlinear Classification
Recommender System
Feedforward Neural Networks
Recurrent Neural Networks
Convolutional Neural Networks
Clustering
Reinforcement Learning
Natural Language Processing

View certificate ↗

June 2022

ONS:
1st Place Datathons 3rd Edition 2022

1st Place at DATATHONS 3rd Edition 2022
Organized by ONS (National Operator of the Electricity System)
Project: Wind Prediction Bias Correction

View certificate ↗

May 2022

AWS Training and Certification:
Machine Learning Learning Plan (35 H)

AWS SageMaker
AWS Rekognition
AWS Polly
AWS Lex
AWS Transcribe
AWS Comprehend
Amazon Elastic Inference

View certificate ↗

November 2021

edX HarvardX:
Data Science Professional Certificate Program

PH125.1x: R Basics
PH125.2x: Visualization
PH125.3x: Probability
PH125.4x: Inference and Modeling
PH125.5x: Productivity Tools
PH125.6x: Wrangling
PH125.7x: Linear Regression
PH125.8x: Machine Learning
Data Science: Capstone

June 2021

edX HarvardX CS50AI:
CS50's Introduction to Artificial Intelligence with Python

Python
pandas
pomegranate
tensorflow
nltk
numpy
sklearn
keras

August 2021

Digital Innovation One:
Banco Carrefour Data Engineer

Python
MongoDB, Neo4j, Cassandra, Redis
Apache Hadoop, Sqoop, Impala, Zookeeper, Hive, Spark
MicroStrategy
Docker
Google Cloud Platform

View certificate ↗

May 2021

Instituto de Gestão e Tecnologia da Informação:
Fullstack Developer (Top Performance)

JavaScript
React
Node.js
Heroku
MongoDB

View certificate ↗

March 2021

Digital Innovation One:
Inter Java Developer

Java
Spring Boot
API REST
JDBC / JPA
JUnit
PostgreSQL / MySQL

View certificate ↗

Machine Learning, LLMs and Knowledge Graphs

Tools and Frameworks

40+ projects on graphs, learning systems and AI

Kodah - SWE-PolyBench Verified (JS/TS/Java)

Kodah - SWE-PolyBench Verified (Python)

Kodah - SWE-bench Lite

Colony Simulator

LLM Playground - Photo Album (Pt.8)

ArXiv Temporal Graph

LLM Playground - Context Window (Pt.7)

LLM Playground - MCP (Pt.6)

LLM Playground - Multi Models (Pt.5)

Graph Component Algorithms

Graph Pathway Algorithms

LLM Playground - SLM (Pt.4)

LLM Playground - EDA (Pt.3)

LLM Playground - Multiagents (Pt.2)

LLM Playground (Pt.1)

Kolmogorov-Arnold Network (KAN)

LLM Improvement with RAG in Knowledge-Graph

NLP Neural Networks

Pokémon Battle Simulator

PySpark ML

Property Rentals

Countries Clustering

JWST Star Detector

Wind Prediction Bias Correction

Heartbeat Abnormality Detector

Digit Recognition App

Digit Convolutional Neural Network

Digit Feedforward Neural Network

Movie Recommendation

Digit Dimension Reduction

Digit Recognition

Iris Species

Polls Regression

Elections Polls

Holy Books Sentiment Analysis

Minesweeper

Tic Tac Toe

Bank Interests

Titanic Prediction

Titanic Data Analysis

Certificates

Hugging Face:Agents Course

Udemy:Graph Theory Algorithms

Activeloop / Towards AI / Intel Disruptor Initiative:Retrieval Augmented Generation for Production with Langchain & LlamaIndex

Neo4j GraphAcademy:Neo4j Graph Data Science Certification

edX IsraelX:Essentials of Genomics and Biomedical Informatics

Google Skillshop:Google Analytics Certification

Neo4j GraphAcademy:Neo4j Certified Professional

Datacamp:Machine Learning Scientist with Python Track (93 H)

Datacamp:Professional Data Scientist

Digital Innovation One:Unimed-BH Data Science

Datacamp:Associate Data Scientist

Datacamp:Data Scientist with Python Track (96 H)

edX MITx 6.86x:Machine Learning with Python-From Linear Models to Deep Learning (150 H)

ONS:1st Place Datathons 3rd Edition 2022

AWS Training and Certification:Machine Learning Learning Plan (35 H)

edX HarvardX:Data Science Professional Certificate Program

edX HarvardX CS50AI:CS50's Introduction to Artificial Intelligence with Python

Digital Innovation One:Banco Carrefour Data Engineer

Instituto de Gestão e Tecnologia da Informação:Fullstack Developer (Top Performance)

Digital Innovation One:Inter Java Developer

Let's talk

Hugging Face:
Agents Course

Udemy:
Graph Theory Algorithms

Activeloop / Towards AI / Intel Disruptor Initiative:
Retrieval Augmented Generation for Production with Langchain & LlamaIndex

Neo4j GraphAcademy:
Neo4j Graph Data Science Certification

edX IsraelX:
Essentials of Genomics and Biomedical Informatics

Google Skillshop:
Google Analytics Certification

Neo4j GraphAcademy:
Neo4j Certified Professional

Datacamp:
Machine Learning Scientist with Python Track (93 H)

Datacamp:
Professional Data Scientist

Digital Innovation One:
Unimed-BH Data Science

Datacamp:
Associate Data Scientist

Datacamp:
Data Scientist with Python Track (96 H)

edX MITx 6.86x:
Machine Learning with Python-From Linear Models to Deep Learning (150 H)

ONS:
1st Place Datathons 3rd Edition 2022

AWS Training and Certification:
Machine Learning Learning Plan (35 H)

edX HarvardX:
Data Science Professional Certificate Program

edX HarvardX CS50AI:
CS50's Introduction to Artificial Intelligence with Python

Digital Innovation One:
Banco Carrefour Data Engineer

Instituto de Gestão e Tecnologia da Informação:
Fullstack Developer (Top Performance)

Digital Innovation One:
Inter Java Developer