Dhruv Lawaniya
I turn messy data into models that actually ship — building end-to-end ML pipelines, from feature engineering to deployment, that hold up in production.
About Me
I'm an ML Engineer and Master of Computer Science candidate at the University of Sydney, focused on building models that go beyond the notebook.
My work spans the full ML lifecycle — data pipelines with Spark and SQL, model development with PyTorch, TensorFlow and scikit-learn, and deployment with Docker and cloud platforms like AWS and GCP. I care about the parts most people skip: reproducibility, monitoring, and making sure a model still works once it's live. I'm currently open to full-time ML Engineer / Data Scientist roles.
Skills & Tools
Programming
Machine Learning & Data Engineering
MLOps, Cloud & Tools
Projects
CIFAR-10 Image Classification Comparison
Head-to-head comparison of PCA+Linear SVM, a tuned MLP ensemble, and a CNN on CIFAR-10 — with full hyperparameter tuning, colour-space experiments, and confusion-matrix/ROC-AUC evaluation. The tuned CNN reached 74.6% test accuracy, roughly double the best classical baseline.
GridPulse
Real-time visualisation of Australia's electricity network — a two-stage pipeline that ingests live power, emissions, price, and demand data from the NEM and streams it over MQTT to an interactive live map, sized by output and colour-coded by fuel type.
ExamGenerator
A one-stop web tool that turns lecture content into exam-style question papers — feed an AI chatbot your lecture materials to generate a structured JSON of MCQ, fill-in-the-blank, short and long answer questions, then upload it to instantly produce a ready-to-use exam.
Experience
- Contributed to a capstone project focused on developing an Agentic AI system for real-time phishing detection and prevention. Involved in researching and evaluating suitable machine learning and AI approaches, contributing to system design, data preparation, model development, and project documentation. Worked collaboratively to translate a real-world cybersecurity problem into a practical AI-driven solution.
- Designed and deployed scalable Scala-Spark data pipelines to process large-scale audit datasets across 100+ enterprise clients, improving processing efficiency and ensuring 100% compliance accuracy.
- Built Python-SQL analytics frameworks to automate anomaly detection and large dataset validation, reducing manual audit effort and improving turnaround time.
- Developed executive-level dashboards using Tableau and Power BI to translate complex analytical outputs into actionable business decisions.
- Collaborated with cross-functional stakeholders to convert regulatory and business requirements into production-ready analytics systems under strict deadlines.
- Worked on data analysis and model-building tasks supporting business decision-making.
Education
- Specialisations: Data Science and AI, Cybersecurity.
- WAM: Distinction.
- CGPA: 9.16
Resume
Want the full picture?
View or download a complete breakdown of my experience, education, and skills.
Get in Touch
Have a role, a project, or just want to say hi? Reach out directly or send a message using the form.
Email dhruvsparee@gmail.com GitHub github.com/DhruvLawaniya LinkedIn linkedin.com/in/dhruvlawaniya