Vikrant Sharma
Contact
Vikrant Sharma

Vikrant Sharma · Adelaide, AustraliaMachine learning,built to hunt threats.

Intrusion detection, threat triage and fraud models that run in production, not in notebooks. Seven systems live, all open source. This dashboard is one of them.

View selected workThe 30 second version

Open to internships and part-time work now

Live hybrid ML intrusion detection dashboard built by Vikrant Sharma

Professional journey

Security floor to ML systems.

Ten months of industry experience across two roles, one in security operations and one in data engineering. That order is the point: I learnt what attacks look like before I started building models to catch them.

  1. Feb 2025 to Aug 2027

    Master of Information Technology

    University of the Sunshine Coast, Adelaide · Adelaide, AU

    • Postgraduate study across data systems, software engineering and cybersecurity research.
    • Capstone: the hybrid ML intrusion detection system featured in the work below, built and deployed while studying.
    • Data Systems
    • Software Engineering
    • Cybersecurity Research

  2. Dec 2024 to Feb 2025

    Data Engineer

    Nagarro · Adelaide, AU

    • Designed and optimised dimensional data models on AWS Redshift and Snowflake for enterprise analytics workloads.
    • Built ETL pipelines turning high-volume transactional data into analyst-ready tables.
    • Delivered weekly under Agile alongside analysts and product owners.
    • Redshift
    • Snowflake
    • ETL
    • Agile

  3. May 2023 to Dec 2023

    Cybersecurity Junior Analyst

    AT SecurDI · Ahmedabad, IN

    • Monitored SIEM platforms and triaged alerts to surface genuine incidents.
    • Authored incident response runbooks adopted by the wider security operations team.
    • Ran OWASP Top 10 web testing and ISO 27001 plus NIST CSF compliance audits across client environments.
    • SIEM
    • Incident Response
    • OWASP
    • ISO 27001
    • NIST CSF

  4. 2021 to 2024

    Bachelor of Computer Applications

    CHARUSAT, CMPICA · Gujarat, IN

    • Undergraduate computing degree. NASA Space Apps Global Finalist and class representative during this period.
    • Computer Applications

Selected work

Three systems, end to end.

Each one is deployed and public. Open a case study for the full write-up, or jump straight to the running demo. Source for everything is on GitHub.

Screenshot of Hybrid ML Intrusion Detection
01 Capstone · Cybersecurity ML

Hybrid ML Intrusion Detection

Challenge
Signature-based tools miss novel attacks, and SOC analysts drown in alerts with no context about what an attacker is actually doing. The job was a detector that catches unknown behaviour and tells the analyst what it means.
Outcome
Detects DDoS, brute force, port scans, web attacks, infiltration and botnet activity across the CICIDS2017 categories. The ATT&CK mapping turns each alert from a raw flag into a triage decision in seconds. Deployed live on Hugging Face Spaces with full source on GitHub.
  • Python
  • PyTorch
  • scikit-learn
  • Scapy
  • Streamlit
  • Docker
Read case study Live demo Source ↗
Screenshot of Credit Card Fraud Detection
02 ML · Imbalanced Classification

Credit Card Fraud Detection

Challenge
The dataset has 284,807 transactions and a 0.17 per cent fraud rate. A naive model scores 99.8 per cent accuracy by predicting that nothing is ever fraud, so accuracy is a trap and the whole problem lives in how you handle the imbalance.
Outcome
A model evaluated on the metric that actually matters for fraud, the precision-recall trade-off at a usable operating threshold, with the full analysis open for review on Hugging Face and GitHub.
  • XGBoost
  • scikit-learn
  • SMOTE
  • Streamlit
Read case study Live demo Source ↗
Screenshot of Cosmic Keys
03 NASA Space Apps 2023 · Global Finalist

Cosmic Keys

Challenge
Planetary datasets are tables of numbers. NASA Space Apps asked teams to make space data accessible to people who will never read a CSV.
Outcome
Selected as a Global Finalist from thousands of teams worldwide. The judges did not read the data, they heard it.
  • Python
  • Streamlit
  • NumPy
  • PIL
Read case study Live demo Source ↗

Also shipped

Technical expertise

What I do, with receipts.

Four domains, each anchored to work you can open and run. No skill appears here without a project or role behind it.

01

Threat Detection ML

Anomaly detection, imbalanced classification and ensemble models for security data, with every alert mapped to MITRE ATT&CK. Most analysts cannot build a model and most ML engineers have never read ATT&CK. I work in the overlap.

Evidence: the IDS capstone and fraud model above.

  • Python
  • PyTorch
  • scikit-learn
  • XGBoost
  • SMOTE

02

Data Engineering

Dimensional modelling, ETL pipelines and warehouse work on AWS Redshift and Snowflake. Production experience turning high-volume transactional data into analyst-ready tables.

Evidence: Data Engineer at Nagarro, AWS Certified Data Engineer.

  • SQL
  • Snowflake
  • Redshift
  • Airflow
  • Pandas

03

Security Operations

SIEM monitoring, alert triage, OWASP Top 10 testing and compliance audits against ISO 27001 and NIST CSF. The operational grounding that tells me what a detection model actually needs to catch.

Evidence: seven months as a security analyst at AT SecurDI.

  • SIEM
  • Wireshark
  • Suricata
  • OWASP
  • NIST CSF

04

Production Delivery

APIs, dashboards and deployment pipelines that put models in front of users. Everything I build ships behind a URL with CI, containers and monitoring.

Evidence: seven public deployments, all with source.

  • FastAPI
  • Flask
  • React
  • Docker
  • GitHub Actions
Languages
Python · SQL · TypeScript · JavaScript · Bash · C
ML & data
PyTorch · scikit-learn · XGBoost · Pandas · NumPy · Airflow · Snowflake · Redshift · PostgreSQL
Security
MITRE ATT&CK · OWASP Top 10 · NIST CSF · ISO 27001 · SIEM · Suricata · Wireshark
Web & cloud
FastAPI · Flask · React · Astro · Docker · AWS · OCI · Cloudflare · GitHub Actions · Linux

Achievements

Verified, not claimed.

Every credential below links to its verification page or original announcement.

NASA International Space Apps Challenge · 2023

Global Finalist

Selected with Team Eklavya from thousands of teams worldwide for AstroSonify, a planetary data sonification engine. The project lives on as Cosmic Keys in the work above.

See the announcement ↗

Further certificates

  • icSoftComp 2022, Intl. Conference on Soft Computing certificate

    icSoftComp 2022, Intl. Conference on Soft Computing

    Springer × CHARUSAT (GUJCOST sponsored) · Dec 2022

  • Class Representative Recognition certificate

    Class Representative Recognition

    CHARUSAT University of Science and Technology · Apr 2023

  • TCS iON Career Edge, Young Professional certificate

    TCS iON Career Edge, Young Professional

    Tata Consultancy Services (TCS iON) · Sept 2023

  • Backend Web Development, Express & Node.js certificate

    Backend Web Development, Express & Node.js

    DevTown × GDSC KIIT × AWS Community Builders · Jan 2023

Off the clock

Three consecutive 2v2 podium finishes in the ASUS ROG Showdown series, 2024. Competitive instinct survives outside work hours.

Resume

The 30 second version.

One page, current as of May 2026. Take the PDF for your ATS or read it in the browser.

Download PDF Read online
  • ML engineer specialising in cybersecurity, Adelaide based
  • Two industry roles: data engineering at Nagarro, security analysis at AT SecurDI
  • AWS Certified Data Engineer Associate, OCI Data Science Professional
  • Seven production deployments, all public, all open source
  • NASA Space Apps Global Finalist 2023
  • Master of IT at UniSC Adelaide, graduating August 2027

Writing

Notes from the field.

All posts →