Skip to content
View rivindu-ashinsa's full-sized avatar

Block or report rivindu-ashinsa

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 is supported. This note will only be visible to you.
Report abuse

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

Report abuse
rivindu-ashinsa/README.md

Rivindu Ashinsa

AI Research & Development Engineer  ·  Data Science  ·  Biomedical ML  ·  Agentic AI

LinkedIn Portfolio Email Instagram


Profile Views


$ whoami

rivindu_ashinsa = {
    "role"      : "AI Research & Development Engineer",
    "company"   : "CognivusLabs",
    "location"  : "Sri Lanka 🇱🇰",
    "domains"   : ["Biomedical ML", "Data Science", "Agentic AI", "Predictive Modeling"],
    "building"  : "AI that bridges research and real-world healthcare impact",
    "philosophy": "Data is the new medicine; AI is the new lab."
}

I build production-grade ML systems for real-time biomedical applications — from ECG arrhythmia detection to SpO₂ forecasting. My work sits at the intersection of rigorous research and engineering reliability.

What drives me:

  • 🏥 Applying AI to healthcare through predictive modeling and signal processing
  • 📊 Solving real-world data science problems — not just academic benchmarks, but models that actually work in production
  • 🧩 Designing interpretable, production-aware systems from research prototypes
  • 🤖 Architecting agentic AI frameworks capable of reasoning and autonomous execution
  • ⚡ Pushing efficient AI to edge devices for real-time health monitoring

🧰 Technical Stack

Core ML & Data Science

Python · SQL · NumPy · pandas · Polars
scikit-learn · PyTorch · TensorFlow
Seaborn · Matplotlib · Plotly

Deep Learning

CNNs · RNNs · Transformers
Time-Series Forecasting
Computer Vision (OpenCV) · NLP

Agentic AI & LLMs

LangGraph · LangChain · Agno
Multi-Agent Systems · ReAct · RAG
Prompt Engineering · Tool Calling

MLOps & Engineering

Git · GitHub · Docker
CI/CD Pipelines · Model Serving
API Development · Edge Deployment

Domain Specialization: Healthcare & Biomedical AI — ECG processing · SpO₂ prediction · Vital signs analysis · Clinical decision support


🔬 Featured Projects

🏥 CognivusMed — Biomedical AI Suite

AI R&D Engineer · CognivusLabs

Real-time biomedical signal intelligence platform.

  • ECG Abnormality Detection — ML models for arrhythmia classification and waveform anomaly detection
  • SpO₂ Prediction Engine — 180-second oxygen saturation forecasting from live signal streams
  • Impact — High-accuracy, low-latency predictions enhancing patient monitoring

🧠 Dementia Prediction System

Team-XPredators

Early-stage dementia risk assessment using longitudinal clinical data.

  • Robust ML models on behavioral and clinical inputs
  • Feature engineering from long-horizon patient records
  • Focus on interpretable outputs for clinical usability

❤️ Heart Disease Prediction Pipeline

Healthcare Analytics

End-to-end cardiovascular risk assessment system.

  • Advanced preprocessing & ensemble modeling
  • Interpretable outputs with Explainable AI techniques
  • Deployable clinical decision support module

🤖 RepoDigest

Open-Source Tool

Intelligent GitHub repository analysis and summarization.

  • Extracts structural insights and documentation patterns
  • Auto-generates comprehensive project summaries
  • Useful for onboarding and open-source discovery

📦 SupplyGenie

OutllierRejects × Idealize

ML-driven supply chain optimization platform.

  • Demand forecasting and inventory intelligence
  • Decision support for logistics workflows
  • End-to-end ML pipeline with production deployment

⚡ EdgeVitalsAI

Edge Computing for Healthcare

Lightweight vital-sign AI for edge hardware.

  • Efficient model architectures for constrained devices
  • Real-time health monitoring with minimal overhead
  • Contributions to open-source edge health tooling

📊 GitHub Analytics

Rivindu Ashinsa GitHub Stats    Rivindu Ashinsa Top Languages
Rivindu Ashinsa Contribution Graph

💡 Research & Problem-Solving Interests

I'm drawn to problems where data science delivers tangible, real-world impact — not just models that perform on paper, but systems that hold up in production.

Area What I'm Working On
📊 Real-World Predictive Modeling End-to-end pipelines that work outside the notebook — robust preprocessing, feature engineering, and deployment-ready outputs
🏥 Healthcare & Biomedical AI ECG, SpO₂, patient monitoring — where model reliability is non-negotiable
🤖 Agentic AI & Multi-Agent Systems Autonomous reasoning, planning, and tool use at scale
🔍 Explainable AI Interpretability for clinical and high-stakes systems
📈 Time-Series & Anomaly Detection Biomedical signal processing and forecasting under real constraints
Edge AI Efficient inference on resource-constrained hardware
🧠 LLMs in Healthcare Clinical NLP, decision support, RAG pipelines
🔐 Federated Learning Privacy-preserving ML for sensitive medical data

I enjoy data science competitions and problem sets that challenge feature engineering, model generalization, and handling messy real-world data. If you have an interesting dataset or a hard prediction problem — let's talk.


🎓 Professional Background

AI & ML Engineering · CognivusLabs Developing and deploying predictive models for biomedical applications. Building validated data pipelines and production-grade AI solutions in cross-functional research teams.

Open-Source Leadership

Organization Focus
Team-XPredators Healthcare AI Research
OutllierRejects Supply Chain & Logistics ML
EdgeVitalsAI Edge Computing for Health Monitoring

Academic Foundation Computer Science undergraduate — integrating theoretical ML foundations with applied research across biomedical, agentic, and edge AI domains.


🤝 Let's Connect

I'm open to:

  • 🔬 Research Collaborations — ML in healthcare and biomedical domains
  • 🧩 Data Science Problems — predictive modeling challenges, messy datasets, real-world ML
  • 🌟 Open-Source Projects — AI/ML tools, health-tech, agentic frameworks
  • 🎤 Speaking / Workshops — AI/data science events and tech meetups
  • 👨‍🏫 Mentorship — Guiding early-stage ML engineers and researchers
📧 Email 💼 LinkedIn 📸 Instagram
ashinsa.rivindu@gmail.com rivindu-ashinsa @rvndq


"Data is the new medicine; AI is the new lab."

Rivindu Ashinsa · AI R&D Engineer · Building intelligent systems that save lives, one algorithm at a time.

rivindu ashinsa · rivindu-ashinsa · ashinsa rivindu · AI Engineer · Data Scientist · CognivusLabs · Sri Lanka

Pinned Loading

  1. EdgeVitalsAI/Cognivus-Labs-Dev EdgeVitalsAI/Cognivus-Labs-Dev Public

    AI-powered real-time health monitoring system that analyzes wearable sensor data at the edge for intelligent vital insights.

    JavaScript 2

  2. RepoDigest RepoDigest Public

    A powerful tool to automatically generate a comprehensive and well-structured README file for your repository, providing a detailed digest of its purpose, contents, and setup instructions

    Python

  3. Team-XPredators/Dementia-Prediction-xpredators Team-XPredators/Dementia-Prediction-xpredators Public

    Machine learning model for predicting dementia risk using non-medical factors only. Includes data cleaning, feature selection, PCA, and classification models (RF, XGBoost, SVM) with explainability.

    Jupyter Notebook

  4. Heart-Disease-Prediction Heart-Disease-Prediction Public

    AI-driven healthcare project aimed at developing predictive models for early diagnosis and health risk analysis.

    Python 1

  5. AI-Powered-Web-site-Audit-Tool AI-Powered-Web-site-Audit-Tool Public

    Lightweight FastAPI application that audits a single webpage and returns factual metrics, compact AI insights, and prioritized recommendations.

    Python 1

  6. rivindu-ashinsa.github.io rivindu-ashinsa.github.io Public

    This repository hosts the Personal portfolio website for Rivindu Ashinsa, showcasing projects, skills, and contact information, built with HTML, CSS, and JavaScript.

    CSS