Vikrant. Get in touch

VIKRANT, ML ENGINEER SPECIALISING IN CYBERSECURITY

Hey,
there

I AM
VIKRANT.

Available for new opportunities Contact
Vikrant portrait

Master of IT student at USC Adelaide. I build machine learning systems that detect threats. Data engineering and full-stack are how I ship them end-to-end. Finishing my degree in Aug 2027 and sitting the AWS Data Engineer Associate exam next month.

Based in
Adelaide, AU
Graduating
Aug 2027
Currently building Hybrid ML IDS, capstone 2026
ML FOR THREAT DETECTION NIDS / HIDS ANOMALY DETECTION MITRE ATT&CK SCIKIT-LEARN / PYTORCH DATA PIPELINES SNOWFLAKE / REDSHIFT ORACLE OCI CERTIFIED ML FOR THREAT DETECTION NIDS / HIDS ANOMALY DETECTION MITRE ATT&CK SCIKIT-LEARN / PYTORCH DATA PIPELINES SNOWFLAKE / REDSHIFT ORACLE OCI CERTIFIED

WHAT I CAN HELP YOU WITH

Services.

ML for cybersecurity is the lead. The other three are how I get the model into production. Real shipped work across all four. Click any live demo below to see the actual code.

01

ML for Threat Detection

My core focus. Building intrusion-detection models that map alerts to MITRE ATT&CK, anomaly classifiers on imbalanced data, phishing analysers that catch what email gateways miss. Most cyber analysts cannot build a model; most ML engineers do not know what ATT&CK is. I sit in the middle.

  • scikit-learn
  • PyTorch
  • XGBoost
  • MITRE ATT&CK

02

Data Engineering

How I feed the models. ETL pipelines, dimensional modelling on Snowflake and Redshift, real-time ingest, observability. Three months at Nagarro shipping production data work.

  • Python
  • SQL
  • Snowflake
  • Airflow

03

Cybersecurity Analysis

How I learned what to detect. Seven months at AT SecurDI doing SIEM triage, OWASP Top 10 web testing, ISO 27001 + NIST CSF audits. The hands-on knowledge that informs every detection model I build.

  • SIEM
  • OWASP
  • NIST CSF
  • ISO 27001

04

Full-Stack Delivery

How I ship the work. Dashboards, APIs, and demos that turn the model output into something humans can actually use. Flask, FastAPI, React, Astro, Docker, Cloudflare Pages, Hugging Face Spaces.

  • Flask
  • FastAPI
  • Docker
  • Astro / React

TOOLS OF THE TRADE

Stack.

Six categories. The tools I reach for on data engineering, machine learning, and security work. If you see something here, ask me about the project where I used it.

01 Languages
  • Python
  • SQL
  • TypeScript
  • JavaScript
  • Bash
  • C
02 Data
  • Snowflake
  • Redshift
  • PostgreSQL
  • SQLite
  • Pandas
  • NumPy
  • Airflow
03 ML & AI
  • scikit-learn
  • PyTorch
  • XGBoost
  • SMOTE
  • Streamlit
  • PIL
04 Web & API
  • Flask
  • FastAPI
  • React
  • Astro
  • Tailwind
  • Pydantic
05 Security
  • MITRE ATT&CK
  • OWASP Top 10
  • NIST CSF
  • ISO 27001
  • Suricata
  • SIEM
  • Wireshark
06 Infra & DevOps
  • Docker
  • AWS
  • OCI
  • Cloudflare
  • Hugging Face
  • GitHub Actions
  • Linux

BY THE NUMBERS

The honest version.

Four numbers anyone can verify from this site, my GitHub, my LinkedIn, or my certification credentials. No rounded-up totals, no padded years.

7
Live projects
all on Hugging Face Spaces with source on GitHub
8
Certifications
led by Oracle OCI Data Science 2025
10
Months of industry work
Nagarro data engineering + AT SecurDI cyber
2027
Graduating
Master of IT, USC Adelaide

EXPERIENCE & SKILLS

Where I've worked.

  1. 2025 topresent

    Master's of Information Technology

    University of the Sunshine Coast (Adelaide Campus) · Adelaide, AU

    Postgraduate IT, focused on data systems, software engineering, and cybersecurity research. Working alongside studies on the projects and demos linked from this site.

    • Data Systems
    • Software Engineering
    • Cybersecurity
    • Research
  2. Dec 2024 toFeb 2025

    Data Engineer

    Nagarro · Adelaide, AU

    Designed and optimised data models for AWS Redshift and Snowflake data lakes. Agile/Scrum delivery, analytics for enterprise clients.

    • Python
    • SQL
    • Redshift
    • Snowflake
    • ETL
  3. May 2023 toDec 2023

    Cybersecurity Junior Analyst

    AT SecurDI · Ahmedabad, IN

    SIEM monitoring, alert triage, incident response runbooks, OWASP Top 10 web/app testing, ISO 27001 + NIST CSF compliance audits.

    • SIEM
    • OWASP
    • NIST CSF
    • ISO 27001
    • Incident Response

RECENT PROJECTS, ALL LIVE, NO MOCKUPS

Things I've shipped.

Every "Live" link below points at a real running deployment. Click any of them. They may take 30 seconds to wake from sleep, then the actual ML / API runs.

  • Screenshot of Hybrid ML Intrusion Detection

    01

    Cybersecurity / ML

    Hybrid ML Intrusion Detection

    My capstone. Real-time intrusion detection for SOC analysts. Every alert is mapped to MITRE ATT&CK so triage takes seconds, not minutes. Detects the attack categories CICIDS2017 covers: DDoS, brute force, port scans, web attacks, infiltration, botnet.

    • Python
    • PyTorch
    • Scapy
    • Streamlit
    • Docker
  • Screenshot of ThreatLens

    02

    Threat Intelligence

    ThreatLens

    Cuts SOC alert triage time. Pre-ranks new CVEs against exploit availability, asset exposure, and MITRE techniques so the alerts that actually matter rise to the top. Signal-to-noise is the whole point.

    • FastAPI
    • NVD API
    • MITRE ATT&CK
    • Pydantic
  • Screenshot of Cosmic Keys (NASA Finalist)

    03

    Data Science / Sonification

    Cosmic Keys (NASA Finalist)

    NASA Space Apps Global Finalist 2023 with Team Eklavya. Makes planetary data audible: latitude, velocity, and temperature become piano notes. An image scanner turns a planet photo into a melody.

    • Python
    • Streamlit
    • NumPy
    • PIL
  • Screenshot of Phishing Detection Platform

    04

    Cybersecurity / ML

    Phishing Detection Platform

    Catches phishing patterns email gateways miss. Header anomalies, hidden payloads, suspicious URLs, sender spoofing. A second opinion for security teams before they click quarantine.

    • Flask
    • SQLite
    • BeautifulSoup
    • Docker
  • Screenshot of Credit Card Fraud Detection

    05

    ML / Data Science

    Credit Card Fraud Detection

    The hardest version of fraud detection: 0.17 percent positive class on 284k transactions. Without the right balance technique, a model hits 99 percent accuracy by predicting nothing. The work is in the threshold, not the algorithm.

    • XGBoost
    • scikit-learn
    • SMOTE
    • Streamlit
  • Screenshot of Log Analyzer Dashboard

    06

    Cybersecurity / SIEM

    Log Analyzer Dashboard

    A SIEM for teams without a Splunk budget. Parses syslog and web access logs, flags brute force + port scans in seconds, maps every alert to MITRE ATT&CK. Spins up on a single VM.

    • Flask
    • Chart.js
    • Regex
    • Docker
  • Screenshot of Password Strength API

    07

    Security / API

    Password Strength API

    Drop-in password check for any auth flow. Returns entropy, breach status (without the server ever seeing the password), and crack-time estimates in under 100 milliseconds.

    • FastAPI
    • HIBP API
    • Docker

AWARDS & RECOGNITION

Wins.

  • ASUS Republic of Gamers × The Sports Club · 2024

    ASUS ROG Showdown, The Sports Club

    Podium

    Three consecutive 2v2 podium finishes across the ASUS ROG Showdown competitive series at The Sports Club, finishing 2nd, 2nd, and 3rd in successive events.

  • NASA · 2023

    NASA Space Apps Challenge, Global Finalist

    Finalist

    Global Finalist in the NASA Space Apps Challenge 2023 with Team Eklavya. The team built AstroSonify, a system that converts planetary data (latitude, longitude, velocity, temperature) into piano tones, with image-scan-to-music capabilities.

    See announcement ↗