Imansha Dilshan@imanshadilshan
2
Events attended
2
Submissions made
Sri Lanka
1 year of experience
About me
I have successfully completed my Internship on Generative AI initiatives, where I gained practical experience in developing and integrating AI-powered solutions for real-world applications. Through this experience, I worked with modern AI technologies, APIs, embeddings, vector databases, and automation workflows while improving my understanding of scalable AI system design. I also have hands-on experience building AI-powered applications and research-focused solutions using modern technologies and frameworks. My work includes developing MERN stack platforms integrated with AI features, experimenting with multimodal AI systems, Retrieval-Augmented Generation (RAG), semantic search, and secure backend architectures. My areas of interest and experience include: Generative AI applications Context-aware embeddings and semantic search AI-powered automation systems Computer vision and multimodal AI Enterprise knowledge systems using vector databases Secure and scalable AI integrations Beyond development, I actively explore innovative AI ideas for hackathons, research projects, and startup-focused solutions, with a strong interest in AI Research & Development.
🤝 Top Collaborators
.png&w=640&q=75)
.png&w=640&q=75)
🤓 Latest Submissions
Zynaptrix Context-Aware RAG (Gemini AI)
Zynaptrix is a context-aware RAG system built around Google Gemini and a modular agent pipeline to convert heterogeneous documents into accurate, professional reports. The backend ingests PDFs and images, uses a captioner, figure-splitter, and table transformer to extract structured content, then creates embeddings and indexes for fast retrieval. An orchestrator agent composes retrieved context with Gemini-powered synthesis; a report-writer agent formats outputs into exportable documents while a safety-critic agent enforces guardrails. The architecture separates ingestion, embedding, retrieval, and generation concerns so teams can add new extractors or retrievers easily. The frontend provides an assistant API and machine-focused workflows for interactive querying and batch report generation. Designed for research and enterprise use, Zynaptrix emphasizes provenance (source links and extracted snippets), multimodal understanding (text, figures, tables), and reproducible report generation for knowledge workers, analysts, and technical communicators
19 May 2026
GeoRescue: Omni GIS Agent
GeoRescue is a multi-agent disaster response platform built for flood emergencies and hackathon-grade real-world impact. It gives first responders and command centers an instant operational view of flood conditions by combining satellite image analysis, live weather-driven flood modeling, road-impact detection, and safe route planning in one workflow. The system uses a CrewAI agent pipeline to coordinate four specialized roles: one analyzes satellite imagery, another triggers live flood intelligence, a third performs geospatial overlay analysis to identify blocked roads, and the last generates a structured incident report. The backend uses FastAPI, GeoPandas, OSMnx, NetworkX, and live weather inputs, while the UI presents interactive maps and clear response outputs. GeoRescue is designed to stay useful even when parts of the stack are unavailable, falling back gracefully to local routing and template reporting. The result is a practical, resilient disaster-response assistant that turns fragmented GIS and AI signals into actionable emergency guidance within seconds.
10 May 2026
