About
Applied AI Scientist at Starbound Space Solutions. PhD researcher at Monash University, exploring generative AI for regulatory compliance.
- Applied AI Scientist
- Starbound Space Solutions
- PhD, Monash University
- Melbourne
I build AI agents and machine-learning systems for safety-critical, regulation-heavy work, and I bring research rigour to production code.
From Swinburne to a Monash PhD
- Bachelor of Computer Science (Artificial Intelligence), Swinburne University of Technology
- Monash University, since Jul 2026
- Starbound Space Solutions
My research explores generative-AI methods that help engineers meet regulatory requirements in safety-critical domains, starting with the Australian space sector. The project is part of the National Industry PhD Program.
The project is a partnership with Starbound Space Solutions, where I work as an Applied AI Scientist, and is supported by a Faculty of Information Technology research scholarship.
Five roles across space missions, drug discovery, factory floors, social platforms and satellite imagery.
CSIRO Manufacturing
Machine Learning Engineer, Feb 2025 to Feb 2026
Pictured with my two supervisors.
- Developed and evaluated an AI application for drug discovery that predicts molecular pharmacokinetic properties to support faster, cheaper drug screening.
- Engineered Graph Neural Network representations and trained multi-task models on HPC with SLURM, placing Top 13 of 370+ global participants in an AI-for-drug-discovery challenge (macro-averaged RAE 0.588).
- Worked across AI, chemistry and bioengineering teams to assess practical applicability and report findings to stakeholders.
Experience
(05)
Starbound Space SolutionsApplied AI ScientistAI-assisted requirements derivation from space mission briefs and ECSS standards.- Developed AI-assisted requirements derivation for space mission briefs and ECSS regulations and standards, capturing 98% of engineering implications and contributing to a team-wide 40% reduction in documentation time.
- Implemented ReqIF 1.2 exports and attribute-gap reports that turn extracted requirements into structured, traceable outputs for engineering review and exchange.
- Deployed on-premises LLM services and strengthened backend reliability, with CI/CD pipelines covering Docker-based verification and automated schema, round-trip and integration tests.
- CSIRO ManufacturingMachine Learning EngineerGraph neural networks that predict pharmacokinetic properties for drug discovery.
- Developed and evaluated an AI application for drug discovery that predicts molecular pharmacokinetic properties to support faster, cheaper drug screening.
- Engineered Graph Neural Network representations and trained multi-task models on HPC with SLURM, placing Top 13 of 370+ global participants in an AI-for-drug-discovery challenge (macro-averaged RAE 0.588).
- Worked across AI, chemistry and bioengineering teams to assess practical applicability and report findings to stakeholders.
FPT SoftwareMulti-modal AI Engineer InternFine-tuned the BLIP multimodal model to read meters on a factory floor.- Fine-tuned the BLIP multimodal model for meter reading in a factory setting using labelled images and question-answer pairs.
- Reached 87% meter-reading accuracy on 500 held-out factory images, 35 percentage points above the pretrained BLIP baseline.
UnibuzFounding Full-stack DeveloperShipped a microservice social platform connecting students across universities.- Shipped a microservice-based social platform connecting students across universities: groups, events, news feed, chat and marketplace.
- Built the interface in React, TailwindCSS and TypeScript, designed the MySQL database and developed REST APIs with Supabase.
AetoskyComputer Vision InternMask R-CNN segmentation of satellite imagery for infrastructure planning.- Labelled satellite imagery in QGIS and configured Mask R-CNN to segment buildings, roads, trees and pools for infrastructure planning.
- Improved model accuracy by 15% by tuning the ROI ratio to 0.5 and the RPN confidence threshold to 0.9.
Generative AI for Regulatory Compliance in Safety-critical Domains
The National Tribune
Monash and Starbound National Industry PhD project
Scholarship offer
Faculty of Information Technology, personal details redacted
Monash student card
Number and barcode removed
Focus
(04)
- AI agent orchestration
- Knowledge representation
- Requirements engineering
- Predictive machine learning
Presenting at CSIRO
Presenting a workflow diagram with my two supervisors at CSIRO Manufacturing, where I worked as a Machine Learning Engineer.
Read the drug-discovery case studyTop 13
of 370+ global participants in an AI-for-drug-discovery challenge
Education
Monash University
Doctor of Philosophy
Jul 2026 - Present. National Industry PhD Program
Swinburne University of Technology
Bachelor of Computer Science (Artificial Intelligence), Professional Degree
Aug 2022 - May 2026. GPA 3.6, top 15% of cohort, scholarship for high academic achievement
Awards
Agentic AI Build Week, GenAI Fund
Awarded around US$160K at the largest buildathon in South East Asia: 2,000+ builders, 10+ enterprise partners and 500+ agentic AI projects across Asia-Pacific.
Golden Key International Honour Society
Member
Recognises the top 15% of students across all disciplines.
Deep Learning and Machine Learning Specialisations
8 courses
Foundations of machine learning and deep learning.
Toolkit
Agents & LLMs
- Claude Code
- ElevenLabs
- FastMCP
- OpenAI Agents SDK
- Azure Foundry
Machine learning
- PyTorch
- Hugging Face
- Chemprop
- RDKit
Backend & infrastructure
- Docker
- PostgreSQL
- Prisma
- NestJS
- FastAPI
- Convex
Languages
- Python
- TypeScript
- Linux
- SQL
- C#
Methods
- AI agent orchestration
- Knowledge graphs
- Embeddings
- Graph neural networks
- Multimodal models
- Metric learning
- Computer vision
- SLURM & HPC
- CI/CD
- Cloud computing
