Welcome
An AI Engineer with intensive skills in Machine Learning and Computer Vision. He
had almost 4 years of experience on different projects that use ML and CV to solve real-world
problems.
His daily job is to build Machine Learning pipelines, and models that improve the accuracy of
various processes. Also, researching, designing, implementing, and deploying scalable computer
vision and machine learning solutions.
Frequently used
Experience
OmiGroup, Vietnam
Software Engineer
Jun 2023 - Present
- ● Filtered and identified candidate resumes using Python and FastAPI for backend processing and Streamlit for user interface, integrating Gemini LLM for accurate data extraction, improving the precision of matching candidates to technology and job position requirements
- ● Streamlined the recruitment process by developing a database with Postgres to store and search candidate information based on user queries, increasing efficiency by 30%
- ● Developed a chatbot system to assess users' health by using GPT-4 API and PaddleOCR to extract text from scanned PDFs and analyze health data, improving recommendation accuracy
- ● Enhanced user experience by deploying the system with Spring Boot, Angular, and AWS, and storing personalized conversation history in MySQL, increasing engagement by 25%
- ● Developed a robust system for scoring recognition on golf scorecards, employing YOLOv7, PyTorch, AWS, and FastAPI. Successfully implemented the solution to meet the needs of over 4000 users on the customer's application
Rikai Mind, Vietnam
Software Engineer
Jul 2021 -
May 2023
- ● Developed workflow automation, scheduled and monitored data and computing workflows with Airflow, reducing the number of manual hand-offs by 2 times per day
- ● Developed an API system to seamlessly retrieve data from the Odoo system and integrate it into Chatfule's chat script, enabling efficient handling of customer inquiries via Messenger on the SIM selling page
- ● Developed a web system and comic application for Japanese customers in the role of Backend Golang, responsible for project maintenance, API development, database updates, and task exchanges via Jira Software
- ● Designed and automated an ETL process, leveraging Apache Airflow to extract real-time weather data from the OpenWeatherMap API, transform the information, and efficiently load it into an AWS S3 bucket
- ● Developed and deployed Web APIs using FastAPI to update the Milvus database and integrated the ChatbotAI model into the website, utilizing Kafka for real-time data processing, effectively supporting over 3,000 users
- ● Utilized Python frameworks such as Scrapy, Selenium, and Beautiful Soup to systematically crawl logo images from design websites, employing PySpark for batch processing to enhance the dataset for training text-to-logo generator models, which improved the performance of AI algorithms
- ● Explored algorithms for Initial Depth Estimation, crafted a Keypoint Detection model (Keypoint R-CNN), and developed a Volume Estimation model. Combined these components to establish a robust system capable of accurately estimating the volume of trash from images
Eastgate Software, Vietnam
Software Engineer
Oct 2020 - Jun 2021
- • Addressed challenges posed by noisy data, trained the model using YOLOv5 OBB, and achieved an impressive 87% accuracy in identifying car serial numbers
- • Augmented and trained invoice data using the MMOCR model, achieving an accuracy of 92% for key information extraction challenges
- • Developed a Searching Palm Line Hand service for a database of 3 thousand hands, achieving remarkable efficiency with a response time of 0.04 seconds/request utilizing Faiss Facebook
- • Utilized ResNet101 with triplet loss to extract features from Palm line hand images, achieving an impressive accuracy of 90%
- • Sped up model inference 2 times by converting the model from Tensorflow, and Pytorch to TensorRT
