This position is offered through the ANU Computing Internship ([COMP3820] /[ COMP4820] / [COMP8830])
**Company**\
Trellis Data Group
Business Unit/Division: ML Research Labs
Trellis Data Group provides cutting-edge AI solutions deployed securely at the enterprise-level, specialising in knowledge mastery, transcription, translation, computer vision, and high-security deployment.
**Project **\
Why do we need faster, higher, and stronger models?
In this research-focused internship project, the intern will be tasked with optimising deep learning models using advanced libraries such as TensorRT. The core objective is to improve the execution speed, efficiency, and scalability of these models. Key activities will include:
Benchmarking Existing Models: Assess current model performance using standard metrics.
Library Exploration: Experiment with various acceleration libraries (e.g., TensorRT) to identify the most effective methods for model optimization.
Code Integration: Implement optimization techniques in existing deep learning projects by modifying open-sourced code bases.
Testing and Evaluation: Rigorously test the optimized models to compare performance improvements, documenting results and methodologies.
Research Documentation: Compile findings and methodologies into a comprehensive research document, contributing to the field’s body of knowledge.
As you guessed, faster means faster inference time, higher means higher throughput, and stronger means increased robustness.
**Required technical skills**\
Programming Languages: Proficiency in Python is essential.
Libraries and Frameworks: Experience with deep learning libraries (e.g., TensorFlow, PyTorch) and specifically with model optimization libraries like TensorRT.
Version Control: Familiarity with Git for version control and collaboration in open-sourced projects.
Computational Skills: Understanding of GPU computing and neural network deployment.
** Special Requirements **\
Police check required.
**Required professional/other skills**\
Analytical Skills: Strong problem-solving skills with a focus on performance metrics and optimization.
Research Ability: Capable of conducting independent research and synthesizing findings effectively.
Communication: Excellent written and verbal communication skills for documenting the research process and outcomes.
Collaboration: Ability to work collaboratively in a team-oriented environment, especially in open-source projects where collaboration is crucial.
**Delivery Mode**\
In-person internship.
**Location**
Canberra, ACT.
**Type of internship**\
Unpaid placement.
**How to apply**\
Applications are invited from students who have already passed the eligibility checks for the Computing Internship courses COMP3820 or COMP4820 or COMP8830. Further information about the Computing Internships can be found on the Computing Internship page.
You can nominate multiple preferred Internship projects and host organisations through the one application form.
The closing date for Expressions of Interest for internship projects is 19th May, 2024. Students who have passed the eligibility checks would have received the application form.