SHARVIL PRABHUDESAI · SYDNEY, NSW, AUSTRALIA

I build AI tools —
and then test whether they actually work.

I’m Sharvil, a Master of Computer Science student at the University of Sydney, specialising in Data Science and AI. I build AI tools, data models, and web applications.

DATA SCIENCE / AI / SOFTWAREExplore the selected work

01 / SELECTED WORK

Built around real problems.

10 projects across AI workflows,
developer tools, modelling, and data systems.

GEOSPATIAL DATA ENGINEERING01
NSW EV CHARGING DATAEvery match, with evidence.
TfNSWABSOpen Charge MapOpenStreetMap
GeoPandasMatchingDuckDB
SAVED RUN24 / 24 validation checks passed

Conceptual spatial data pipeline

Geospatial Data Engineering

ChargeAtlas NSW

A reproducible pipeline that connects NSW EV charging records with regional boundaries and external station data, keeping the evidence behind every match.

PythonGeoPandasDuckDBpandasRapidFuzzJupyter
AI IDEA EXPLORATION02
ONE ROUGH IDEASeveral ways forward
Frugal founderDomain expertContrarian
COMPARE · CRITIQUE · SHORTLISTA tournament of ideas
Jev
Pairwise judging
Claude critics
Strong objections

Conceptual idea-evaluation workflow

AI Workflows & Decision Support

IdeaJoust

Turn a rough idea into a shortlist of competing approaches — judged in a pairwise tournament, attacked by AI critics, and decided by you.

PythonSQLiteLLM agentsClaude CodeJev APIGitHub Actions
View case studyWatch explainer Code available on request
TRACELENS / DASHBOARD03
TraceLens workspace dashboard with project navigation, usage totals, and validation findings from its bundled synthetic demo data.

Real dashboard · synthetic demo data

Developer Analytics & Project Memory

TraceLens

Offline Codex analytics with traceable usage, validation evidence, and project memory for the next session.

PythonSQLiteHTMLJavaScriptCodex
View case studyCode available on request
WAKELINE / DASHBOARD04
Wakeline dashboard with Claude Code session counts, separate token categories, and estimated costs generated from synthetic demo logs.

Real dashboard · synthetic demo data

Developer Tools & Analytics

Wakeline

Local analytics for Claude Code: understand sessions, tokens, estimated costs, and coding friction.

PythonSQLiteJavaScriptClaude CodeGitHub Actions
View case studyWatch explainer Code available on request
PRODUCT DISCOVERY05
AmazonCromaReliance
PRODUCT DISCOVERYTwo recommendation paths
Personalised
Content + collaborative
Trending
Popularity model

Conceptual recommendation flow

Web & Recommendation Systems

Price Insights & Recommendation System

Cross-retailer search, price tracking, and product discovery through hybrid recommendations.

Next.jsTypeScriptPythonFlaskMongoDBPuppeteerCheerio
DATABASE ENGINEERING06
5
Core relational tablesAccounts, tracks & reviews
Pythonpsycopg2PostgreSQL
MY CONTRIBUTIONUser & review operations

Conceptual data-access flow

Database Engineering

Sydney Music Database System

A music catalogue with role-based accounts, track reviews, and a Python data-access layer.

PythonPostgreSQLpsycopg2Flask
PREDICTIVE MODELLING07
SOURCE RECORDS101,000Flight records
PREPARED DATASET93,380Completed flights
3 classifiers compared14 pre-flight predictors
> 15 min

Conceptual modelling workflow

Machine Learning

Flight Delay Prediction

Comparing three classifiers using pre-flight features to predict arrival delays over 15 minutes.

Pythonpandasscikit-learn
ALGORITHM DESIGN & ANALYSIS08
ALGORITHM DESIGNOne problem. Different strategies.
Theory → implementation → analysisDivide & conquer · Greedy choices
Dynamic programming · Backtracking
C examples11 documents24 pages of notes

Conceptual learning flow from algorithm notes to C examples and complexity analysis.

Algorithm Design & Analysis

Modern Algorithms & Strategies

Handwritten theory and documented C examples connecting five algorithm design strategies with searching, sorting, optimisation, and string problems.

CDivide & conquerDynamic programmingGreedy algorithmsBacktracking
OPERATING SYSTEMS & SYSTEMS PROGRAMMING09
OPERATING SYSTEMSTrace every decision.
C lab programs · POSIX process examplesSchedule CPU time → serve disk requests
Replace pages → inspect resource safety
27 programs6 core areas9 lab documents

Conceptual overview of scheduling, memory, and resource-management lab exercises.

Operating Systems & Systems Programming

Operating System Algorithms

A collection of 27 C lab programs exploring CPU and disk scheduling, memory management, deadlocks, process creation, and buffer behaviour.

CPOSIXCPU schedulingPage replacementDeadlock handling
AI FOUNDATIONS & SEARCH10
AI FOUNDATIONSFrom search to neural logic.
Python · NumPyExplore a frontier → compare states
Weighted inputs → threshold decisions
SearchPuzzlesRoutesLogic gates

Conceptual overview of search, puzzle states, route selection, and threshold-neuron logic.

AI Foundations & Search

Artificial Intelligence Basics

Eighteen Python lab programs exploring graph search, state-space puzzles, greedy route selection, and neural logic gates, supported by worked examples.

PythonNumPyGraph searchHeuristicsThreshold neurons

02 / A LITTLE ABOUT ME

Curiosity, with
something to show.

I’m a computer science postgraduate based in Sydney, with an interest in how data becomes something useful: a prediction, a recommendation, or an application that helps people make a decision.

My projects connect the analytical and engineering sides of that process, from preparing flight data for a classifier to building relational databases and integrating a price comparison platform. I’m continuing that work through my Master of Computer Science, majoring in Data Science and AI, at the University of Sydney.

MY TOOLKIT

Programming

Python Java C C++ JavaScript TypeScript PHP

AI & LLMs

LLM agents and multi-agent pipelines LLM evaluation (LLM-as-judge) Prompt engineering Recommendation systems Claude Code

Statistics & evaluation

Experiment design Pre-registered hypothesis testing Cross-validation Bradley–Terry ranking

Data science

pandas scikit-learn Data preprocessing Feature engineering Predictive modelling Data visualisation

Web development

HTML CSS React Next.js Flask OutSystems

Databases

SQL PostgreSQL MySQL MongoDB SQLite

Tools

Git GitHub GitHub Actions (CI) pytest Figma

03 / EXPERIENCE

Building for
business.

INTERNSHIP

Project Intern

Tangentia India Technologies Pvt Ltd

Panaji, Goa, India

July 2023 – September 2023

  • Developed B2C and B2B2C applications using OutSystems to support business and logistics operations.
  • Optimised Procure to Pay and Order to Cash workflows supporting purchasing, order processing, and business transactions.
OutSystemsB2CB2B2C

04 / EDUCATION & RECOGNITION

A foundation to build on.

01August 2025 – expected July 2027

University of Sydney

Master of Computer Science

Majoring in Data Science and AI

Sydney, NSW, Australia

82.37 / 100 WAMDistinction

RECOGNITION

  • Sydney Scholars India Scholarship
  • Vice-Chancellor’s International Scholarship
02August 2020 – July 2024

Agnel Institute of Technology and Design

Bachelor of Engineering in Computer Science

Assagao, Goa, India

9.51 / 10 CGPAA+

RECOGNITION

  • Second rank in both third and final year
  • Best Student Award · 2024 graduating cohort

05 / RESEARCH

Thinking beyond
the build.

Co-author of a conference publication on hybrid and popularity-based recommendation systems.

CONFERENCE PUBLICATION · 2024

Survey on Recommendation System Combining Hybrid and Popularity Based Approach

S. Bhogan, S. Naik, S. Gaonkar, S. Prabhudesai, G. Dhond, K. Fernandes.

2024 2nd International Conference on Intelligent Data Communication Technologies and Internet of Things (IDCIoT)

Bengaluru, India · pp. 1573–1578

Read the publication (opens in a new tab)

DOI: 10.1109/IDCIoT59759.2024.10467802

06 / LET’S CONNECT

Have an idea?
Let’s talk.

For a conversation about data, intelligent applications, or a project we could work on together.