Adeesha Perera
A data science undergraduate and machine learning engineer who builds AI-powered systems using RAG, NLP, AI agents, and modern web tech.
Focused on bridging the gap between complex AI systems and practical user applications through data science, MLOps, cloud engineering, and automation-driven solutions.
Featured Projects
Production-grade fraud detection with CatBoost, Feast feature store, Kafka streaming, and automated MLOps — sub-15ms latency.
An interactive RAG-based assistant that answers questions about my skills and experience using local documents.
Enterprise spam detection with DistilBERT, FastAPI, Streamlit, deployed on AWS SageMaker with Terraform IaC.
Automated data pipeline using EventBridge, Lambda, Athena, Glue, and QuickSight for serverless analytics.
Education
Currently pursuing — Expected 2027
Certifications
AWS Machine Learning Associate
Amazon Web Services
Leadership & Achievements
Karate Black Belt
1st Dan — discipline, focus, and years of dedication
Cadet Sergeant
Leadership, teamwork, and responsibility in a structured environment
Tech Stack
I build with a diverse set of tools, but here's the core stack I work with most:
Writings & Blogs
I write about building production ML systems, RAG architectures, MLOps pipelines, AI agents, and cloud-native infrastructure. Deep dives into real projects — from feature stores and model serving to streaming pipelines and automated retraining.
Read on MediumLibrary
*and many more — these are just some of my favorites
Thing about me
Beyond code and AI systems, I find balance in exploring new technologies and understanding how things work at their core. My curiosity drives me to constantly learn, experiment, and push the boundaries of what's possible with modern tech.
I believe the best products come from people who are genuinely curious. It's the unique combination of technical depth and human perspective that allows us to create technology that actually resonates with users.