Skip to content
View rohan2107's full-sized avatar

Highlights

  • Pro

Block or report rohan2107

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
rohan2107/README.md

Hi, I'm Rohan Anthony

Software Engineer (ex-IBM) • MSc Data Science @ University of Bristol
Backend • Systems • Full-Stack • Applied ML

I’m a software engineer with experience building distributed systems, secure C++ agents, backend services, and cloud-deployed machine learning pipelines.

I enjoy working across the stack — from low-level systems and networking to Python APIs, DevOps, and applied ML — with a strong focus on correctness, clarity, and maintainability.


🚀 Featured Projects

🏋️ Gym-Bro — Full-Stack Fitness App

A personal fitness tracking application focused on simplicity and consistency.

  • FastAPI backend with REST APIs
  • React-based frontend (PWA)
  • Workout and meal logging
  • Offline support via service workers
  • Experimental ML-based food / nutrition estimation (in progress)

📌 Repo: https://github.com/rohan2107/gym-bro


🛡️ Sentinel — Windows Security Agent (C++)

A Windows posture and compliance agent.

  • osquery-based posture checks
  • SQL + Lua evaluation engine
  • Policy versioning and local persistence
  • Secure MQTT communication (TLS / mTLS)
  • SHA-256 deduplication
  • SQLite-backed storage

📌 Repo: https://github.com/rohan2107/Sentinel


🤖 Fraud Detection API — ML Service

An end-to-end machine learning service exposed via a REST API.

  • Isolation Forest–based anomaly detection
  • Feature preprocessing and evaluation
  • FastAPI /predict endpoint
  • Dockerised deployment

📌 Repo: https://github.com/rohan2107/fraud-detection


🔧 Tech Stack & Skills

Languages
C++, Python, Java, Go, SQL

Backend & Systems
FastAPI, REST, OpenAPI, MQTT, Kafka, TLS/mTLS, SQLModel

AI / ML
scikit-learn, Pandas, NumPy, Random Forests, Elastic Net, cross-validation

Cloud & DevOps
Docker, Kubernetes, AWS (EC2, ASG, SQS, S3), GitHub Actions, Jenkins

Databases
PostgreSQL, SQLite

Other Tools
osquery, Lua, Git, R (tidyverse, caret)


💼 Experience

Software Engineer — IBM (2023–2025)

  • Built secure C++ communication modules for a Windows agent, improving throughput by ~30%
  • Designed a SQL + Lua posture evaluation engine used across 10k+ devices
  • Developed Java and Go microservices, including a Kubernetes Operator
  • Automated CI/CD pipelines (Jenkins, GitHub Actions), reducing deployment time by ~40%

Software Engineering Intern — IBM Cloud (2023)

  • Worked on Java / Node.js microservices for billing and orchestration
  • Implemented scheduled jobs and authored Swagger / OpenAPI documentation

🎓 Education

MSc Data Science — University of Bristol (2025–2026)

  • Built ML pipelines in R (Elastic Net + tuned Random Forest, F1 = 0.787)
  • Deployed an AWS micro-application using an SQS-driven Auto Scaling Group

B.Tech, Electronics & Communication Engineering — MIT Manipal (2023)


📫 Connect

Pinned Loading

  1. fraud-detection fraud-detection Public

    End-to-end fraud detection pipeline using Isolation Forest, feature engineering, and FastAPI deployment with a real-time /predict endpoint.

    Python

  2. gym-bro gym-bro Public

    A full-stack fitness tracking PWA with AI meal analysis, offline support, and a clean mobile-first UI. Built with FastAPI, React, and service workers.

    TypeScript

  3. Sentinel Sentinel Public

    Rule based risk scoring engine for windows

    C++