AI Engineer with 1+ years in LLM-powered agent systems & FastAPI microservices
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Software Engineering graduand and AI enthusiast with 1+ years of hands-on experience at CodeGen International. Contributed to the development and deployment of LLM-powered agent systems and FastAPI microservices. Strong foundation in Python, machine learning pipelines, and delivering production-ready features.
University of Lancashire, UK
BSC (HONS) SOFTWARE ENGINEERING · Software Engineering
August 1, 2022 – June 30, 2026
Lyceum International School, Sri Lanka
GCE Advanced Level (London)
June 1, 2010 – May 31, 2022
CodeGen International Pvt Ltd
Trainee Software Engineer - R&D Team
August 1, 2024 – September 1, 2025
Colombo, Western Province, Sri Lanka
Automated Data Retrieval & MCP Integration Pipeline
March 1, 2026 – March 1, 2026
Engineered a dynamic, automated workflow using n8n and MCP to extract and process Google spreadsheet data. Established a scalable, future-proof foundation for bidirectional data synchronization, effectively eliminating manual data handling overhead.
View ProjectFacial Expression Recognition CNN
March 1, 2026 – April 1, 2026
Built and fine-tuned an EfficientNet-B2 model for facial emotion recognition utilizing transfer learning, while optimizing hyperparameters via Optuna Bayesian optimization. Engineered a resilient computer vision pipeline using Haar Cascades and CLAHE for preprocessing unconstrained images, and resolved critical dataset imbalances by implementing Focal Loss. Achieved 64.5% accuracy on the highly noisy FER-2013 dataset and conducted zero-shot cross-domain testing on the JAFFE dataset to rigorously evaluate real-world domain shift.
View ProjectEvent-Driven Microservices Platform
January 1, 2026 – March 1, 2026
Engineered a distributed microservices architecture using Node.js and Express, orchestrating highly scalable, asynchronous inter-service communication via RabbitMQ message brokers. Implemented secure OIDC/Google Authentication for protected moderation workflows and successfully managed distributed data persistence across dual databases to ensure high availability. Streamlined infrastructure operations by designing an API Gateway for efficient request routing, utilizing Terraform for Infrastructure as Code (IaC) provisioning, and automating seamless cloud deployments to Azure via GitHub Actions.
Facial Expression Recognition using SVM
December 1, 2025 – January 1, 2026
Investigated demographic overfitting in emotion recognition by training SVM models on Western (CK+) and Asian (JAFFE) datasets using HOG features. Optimized image preprocessing with CLAHE for illumination invariance and utilized Cost-Sensitive Learning to resolve class imbalances.
Real-Time Embedded Temperature Control System Simulation
December 1, 2025 – February 1, 2026
Designed a concurrent simulation of a multi-zone industrial temperature control system. Implemented a thread-safe Physics Engine using Java's ExecutorService to manage parallel sensor updates and actuator adjustments. Utilized ReentrantReadWriteLocks and Atomic variables to prevent race conditions during real-time state mutations.
View ProjectData-Centric ML Architecture for Trustworthy Sepsis Prediction
October 1, 2025 – April 1, 2026
Engineered an end-to-end machine learning pipeline to predict early sepsis, extracting 64 optimized features within a strict 48-hour baseline window to capture physiological momentum and neutralize historical EHR data bias. Trained a robust XGBoost predictive model supported by a decoupled data pipeline, achieving an AUROC of 0.902 and an AUPRC of 0.520 on highly imbalanced clinical data while mathematically anchoring the threshold to a life-saving 85% recall boundary. Established strict clinical trust and algorithmic fairness by achieving an Equal Opportunity Difference (EOD) < 0.1 across patient demographics, translating the complex predictive logic into interactive SHAP dashboards for transparent medical insights.
Data or Specimens only Research
CITI Program
June 1, 2026 – Present
GraphRag Essential Training
June 1, 2026 – Present
AI Fundamentals
June 1, 2026 – Present
Cultural Fit Analysis
The candidate's academic background and project diversity, including contributions to a live corporate hiring system, indicate a proactive and engaged approach to learning and application. Their involvement in team-based activities and mentoring suggests a collaborative spirit. The range of technologies and concepts explored across projects (AI/ML, microservices, embedded systems, cloud) demonstrates intellectual curiosity and a willingness to tackle varied challenges, which aligns well with an innovative R&D environment. However, the experience is primarily academic and internship-level, which might require some ramp-up in a fast-paced industry setting.
Soft Skills & Operational Fit
The candidate demonstrates strong problem-solving skills through complex project implementations (e.g., resolving data imbalance, achieving algorithmic fairness). Their experience mentoring interns suggests leadership potential and a collaborative mindset. The involvement in various extracurricular activities (basketball captain, scrabble captain, senior prefect) indicates discipline, teamwork, and organizational skills. The ability to work on diverse projects, from real-time embedded systems to AI recruitment platforms, suggests adaptability and a broad technical interest.