Larry Long Email & Phone Number

Toronto, On, Ca

Ontario, Canada

- Constructed LLM prompts using modified-RAG, improving end-to-end pipeline accuracy by over 90%- Researched opensource LLMs such as Llama 3, Qwen, mistral-nemo. Proposed a fine-tuned T5 model and presented comparison results to HQ- Implemented Python training code to fine-tune models through hyperparameter tuning; Plotted training graphs to identify optimal training epochs to reduce GPU training costs- Fine-tuned Bert and E5 transformer models to support sentence similarly matching; Prepared and modified Bert and e5 training datasets to achieve 95% training accuracy

Ontario, Canada

- Verified performance improvements using concourse pipeline, memtier benchmark, flame graphs, CPU and memory usage- Performed Redis and Keydb functional regression testing through single, master-slave, cluster, multi-slave deployment scenarios in Linux, EulerOS, and docker environments- Documented step by step testing and error reproduction steps through release test guides and GitLab issues; - Prepared weekly testing reports and performed comparative analysis of current version release improvements- Implemented Java, JavaScript RedisJSON client unit test suites; verified using AWS JSON command reference

Ontario, Canada

- Organized weekly email and discord announcements for club activities across the 2023-2024 school year- Taught a workshop on current market news including topics such as quantitative easing and monetary policy; reviewed 2023 stock market returns and provided insights on future BoC/ Federal Reserve rate cut decisions

Toronto, Ontario, Canada

- Designed a remote controllable rover vehicle capable of operating in rural, unknown terrain and competed in the Canadian International Rover Challenge (CIRC); - Operated rover drive control and sensor integration with Jetson Xavier nx using ROS on Ubuntu 18.04.

Toronto, Ontario, Canada

• Optimized repetitive data diagnostic code into function calls for faster data quality checking • Implemented extension to generate a table image view of any multi-independent test variables compared with any multiple test variable for convenient data research entries

Toronto, Ontario, Canada

- Developed project requirements for the electrical horn and lighting system as well as implementing Ardrino code for motor control.

Collaborated with NewSpark Media to develop a product launch strategy to list NewSpark’s AI tagging technology onto the Appsumo software marketplace with the goal of increasing revenue.

Robotics And Mechatronics Minor

Engineering Business Minor

Bachelor Of Applied Science - Basc, Computer Software Engineering