Help & Supportcareers@gravityer.com
AI Engineer with less than a year in LLM Application & Computer Vision
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Computer Engineering graduate from the Faculty of Engineering, University of Ruhuna, specialising in LLM application engineering, RAG systems, and computer vision. Built and fine-tuned language models for CodeShield project which was selected as a FYP Arena Finalist 2025. Hands-on professional experience deploying AI-powered systems at CodeGen International using LangChain, LangGraph, and vector databases. Seeking an AI/ML engineering role to contribute to production-grade, real-world intelligent systems.
University of Ruhuna
BSc (Hons) · Computer Engineering
August 1, 2021 – June 30, 2026
Central College, Anuradhapura
GCE Advanced Level · Physical Science
N/A – May 31, 2019
CodeGen International (Pvt) Ltd
Trainee Software Engineer
July 1, 2024 – January 1, 2025
Colombo, Western Province, Sri Lanka
Kafka Order Processing System
September 1, 2025 – November 1, 2025
Built a fault-tolerant event-driven pipeline with a Dead Letter Queue (DLQ) pattern, automatic retry with exponential backoff, and real-time running order price averages. Demonstrated production-grade Kafka patterns (Avro schema registry, consumer groups, partition offsets) applicable to high-throughput e-commerce and fintech systems.
View ProjectFootball Analytics Computer Vision System
September 1, 2025 – October 1, 2025
Implemented real-time multi-object tracking using YOLOv5 + ByteTrack to track 22+ players and referees at 30 FPS across 90-minute match videos. Developed automated team classification using K-Means clustering on jersey colour histograms, eliminating manual annotation and correctly classifying teams with greater accuracy. Built an end-to-end video processing pipeline converting raw match footage into structured per-frame player position analytics.
View ProjectEduStream - Educational Course Subscription Platform
July 1, 2025 – September 1, 2025
Designed a microservices architecture with isolated services for authentication, course management, and subscriptions, each independently deployable via Docker Compose. Configured Nginx as an API Gateway for request routing, load balancing, and service abstraction, eliminating direct service exposure. Implemented JWT authentication and role-based access control (RBAC) supporting three distinct user roles: students, instructors, and administrators.
View ProjectCodeShield - AI Code Security Agent
March 1, 2025 – January 1, 2026
Architected the full AI pipeline: GitHub repository ingestion, chunking, Qdrant vector indexing, LangGraph agent orchestration, remediation suggestion generation. Fine-tuned low-parameter Large Language Models on custom code vulnerability datasets, achieving 92% detection accuracy. Built a real-time Next.js frontend using Server-Sent Events (SSE) to stream vulnerability reports token-by-token as the LLM generates them, reducing perceived wait time.
View ProjectSpam Email Classifier
March 1, 2024 – June 1, 2024
Trained LSTM-based deep learning model for binary spam classification achieving 97% accuracy on benchmark dataset of 10,000+ labeled emails. Built text preprocessing pipeline including tokenization, stop-word removal, stemming, and TF-IDF vectorization. Developed Flask REST API for model inference.
iReport - News Reporting Website
November 1, 2023 – June 1, 2024
Implemented an LSTM-based next-word prediction feature embedded in the journalist article editor, reducing average drafting time for repeat users. Built responsive multi-role UI components in React, Redux, and TypeScript for journalists, editors, and administrators.
View ProjectFYP Arena Finalist 2025
University of Ruhuna
January 1, 2025 – Present
Spirit of the Month
AIESEC Sri Lanka
October 1, 2022 – Present
Cultural Fit Analysis
The candidate's diverse range of projects, from AI code security to educational platforms and computer vision, indicates a broad interest and adaptability. The academic and personal projects, combined with a trainee software engineer role, show initiative and a proactive approach to learning and applying new technologies. The volunteering experience further suggests a team-oriented and engaged individual, contributing positively to cultural fit.
Soft Skills & Operational Fit
The candidate's project descriptions and work experience highlight collaboration in Agile teams, contribution to API design reviews, and enforcement of SOLID principles, suggesting a good operational fit. Involvement in volunteering roles like Team Lead and Vice President indicates leadership potential and teamwork skills. The detailed project descriptions demonstrate good communication of technical concepts.