Tam Nguyen

Lead Machine Learning Engineer

Employment Hero

Jun 2024 - Current (2 years 5 months)

• Create multi-agent systems with hierarchical supervisor using ReAct, tool calling for synergizing reasoning.
• Leverage adaptive RAG architecture for query analysis to route across sub-agents and self-corrective RAG
• Implement a long-term memory including semantic, episodic and procedural types in hot path, background.
• Build synthetic datasets, dynamic few-shot examples and evaluation pipeline for the RAG application.

Senior Machine Learning Engineer

Employment Hero

Jul 2023 - Jun 2024 (1 year)

• Implement an advanced RAG multi-vector retriever, including parent-child document, summary, hypothetical queries and self-query to increase retrieval accuracy while retaining the contextual information.
• Use different query transformation methods to improve retrieval, such as rewrite-retrieve-read, step-back prompting, condensed/standalone question, multi-query retrieval and RAG-fusion.
• Implement re-ranking as a second stage to rank retrieved documents using relevance scores.
• Use multimodal LLM to produce summaries from images, then pass them to the LLM for answer synthesis.
• Store long-term historical chat and user feedback for analytics and reinforcement learning (RLHF).
• Plan the LLM RAG evaluation strat

Senior Data Engineer

Global Fashion Group

Jul 2020 - Jun 2023 (3 years)

Join in Global Business Intelligence (GBI) and Data Lakehouse for data science projects. Design data architecture and build data pipelines, data lake and data warehouse for GBI:
• Utilize Apache Airflow for the workflow management and job schedulers.
• Use transient EMR clusters and Glue jobs for data transformation with Spark written in PySpark.
• Store data in S3 data lake with parquet format, partitions and snappy compression for further data analysis by Athena with metadata, partitions and indices are managed by Glue Data Catalog.
• Load cleansed data into Redshift data warehouse with dimensional model and flattened model.
• Create large external tables and query the S3 data by Redshift Spectrum.
• Import c

Data Engineer

Lazada Group

Jan 2018 - Jun 2020 (2 years 6 months)

• Transform business information into models (conceptual, logical and physical) and manage models.
• Manage Alibaba Cloud Data Lake Analytics, Data Warehouses MaxCompute for Business Intelligence
• Build data pipeline with Alibaba Cloud Data Integration and E-MapReduce solutions, write queries and run ETL/ELT jobs to make data accessible in required formats.
• Use Apache Spark in ingesting, filtering, sorting, joining and transforming data sets based on business requirements to support data warehouse development.
• Build the data validation, reconciliation and metadata management components to support the data governance and information management processes.
• Provide operational support for the Data Platform produc

Data Engineer

Pentalog

Jul 2015 - Dec 2017 (2 years 6 months)

• Design database schema for both OLTP, OLAP systems by using ER Diagram, Toad Data Modeler.
• Create OLTP schema: tables, relationships between them, making suitable and effective indexes.
• Create OLAP schema by following Kimball DWH architecture with Star and Snowflake schemes.
• Build data pipeline for both batch and stream processing by utilize Cloud Platform (AWS and GCP).
• Write RESTful APIs with Python and related frameworks, database procedures for backend layer.
• Build high availability solution for OLTP DB system with replication and load balancing solutions.
• Some projects have been done:
• Building Twitter crawler and analysis system for Leading Influencer Platform customer.
• Use Pyth

About

Detail

Timeline

Résumé

Tam Nguyen

About

Detail

Behavioral traits