
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Fullstack Engineer with 2+ years in LangGraph & Agentic AI
Full-Stack AI/ML Engineer with 2+ years of experience designing, shipping, and scaling AI SaaS products and autonomous AI agents. Hands-on experience developing production-ready LangGraph agents, Agentic RAG pipelines (Pinecone, ChromaDB), and FastAPI microservices. Enthusiastic about continuous learning, with practical skills in prompt engineering, evaluation frameworks, and end-to-end MLOps using Docker and AWS, GCP. I am a passionate Software Engineer specialising in AI/ML, with a strong foundation in designing and delivering production-grade Agentic AI systems, LLM orchestration, and enterprise-ready backend services. I thrive in fast-moving environments where engineering rigour and applied AI innovation intersect. I am eager to contribute to a collaborative, cross-functional team and continue growing at the forefront of the AI engineering landscape.
University of Ruhuna
BACHELOR OF SCIENCE (GENERAL) · PHYSICAL SCIENCE STREAM
January 30, 2020 – November 7, 2024
BOUTIQUE REPUBLIC
SOFTWARE ENGINEER
April 1, 2025 – Present
Colombo, Western Province, Sri Lanka
BOUTIQUE REPUBLIC
JUNIOR SOFTWARE ENGINEER
April 15, 2024 – March 31, 2025
Colombo, Western Province, Sri Lanka
BAN-WORLD
FULL STACK INTERN
March 4, 2024 – April 14, 2024
Colombo, Western Province, Sri Lanka
Kapru - Multi-Model Agentic Shopping Assistant (Kapruka Agent Challenge)
June 1, 2026 – June 30, 2026
A trilingual conversational AI shopping agent built on Kapruka's live MCP platform, handling end-to-end discovery to checkout with real order creation. Built for the Kapruka Agent Challenge. Built a single-agent, multi-model system on LangGraph with a Plan->Reason->Act->Reflect loop, routing operational reasoning to Claude Sonnet and multilingual response generation to Gemini based on each model's strengths. Integrated Kapruka's MCP server (Model Context Protocol) for live product search, delivery quotes, and real order creation, with concurrent image enrichment across results. Implemented auto-detected trilingual support (English, Sinhala, Tamil, and romanized variants) with the agent mirroring the user's exact language and style. Added real-time SSE streaming of the agent's reasoning steps to the frontend, with dynamic status labels reflecting live state. Hardened against prompt injection (system-prompt extraction, role hijacking, price manipulation) with code-built orders and real-source pricing protecting the transaction path.
Machine Learning A-Z: AI, Python & R + ChatGPT Prize [2025]
Udemy
April 6, 2025 – Present
The Complete Python Developer
Udemy
February 12, 2024 – Present
Cultural Fit Analysis
The candidate's project diversity, ranging from AI shopping assistants to content generation platforms and MarTech products, demonstrates adaptability and a broad interest in applying AI across different domains. The continuous learning mindset, evidenced by certifications and explicit mention in the 'About Me' section, indicates a strong cultural fit for an evolving tech landscape. The experience in both individual project leadership and contributions to larger systems suggests a balanced approach to teamwork and independent work. The target role of 'Fullstack Engineer' aligns well with the candidate's demonstrated full-stack AI/ML development experience.
Soft Skills & Operational Fit
The candidate's resume highlights a passion for AI/ML and a desire to contribute to collaborative, cross-functional teams, suggesting a good operational fit for dynamic tech environments. The detailed project descriptions indicate strong problem-solving skills and an ability to tackle complex technical challenges, such as prompt injection hardening and multi-model routing. The focus on 'engineering rigour' and 'applied AI innovation' aligns with a high-performance culture.