Ahamed Shahmi
Python Backend Engineer and AI Engineer with hands-on experience building AI-powered backend services, LLM-integrated systems, and RAG pipelines for real-world applications. Strong background in semantic retrieval, embedding, vector databases, and agentic workflows using modern AI frameworks. Experienced in designing production-grade APIs, optimizing inference performance, and deploying scalable, cloud-ready AI systems with a focus on accuracy, reliability, monitoring, and reproducibility.
Experience
Predictive Maintenance Engineering Intern
InvoTech Holdings (Pvt) Ltd
Supported AI experimentation and data analysis for smart solar PV system using Python-based workflows. Developed and tested machine learning models for time-series data analysis, anomaly detection, and energy output prediction. Assisted in data cleaning, data preparation, and data transformation for sensor-based datasets.
Education
University of Ruhuna
Bachelor of Science (Honors)
OGPA: 3.3/4.0 (Second Class – Upper Division). Relevant Coursework: Intelligent System Design, Machine Learning, Object-Oriented Design Patterns, Data Structures and Algorithms, Databases.
Skills
PROJECTS
Multi-Modal AI System for Produce Freshness Detection
Final Year Undergraduate Project
Designed and deployed a real-time multi-modal AI System integration CNN-based image analysis with gas sensor, NIR, and metadata features using early fusion. Achieved 93.42% accuracy with 113 ms interference latency on a portable edge device, outperforming image-only and sensor-only baselines across Accuracy, Precision, Recall, and F1-score. Validate system through real-world field testing in a vegetable shop environment, receiving positive usability feedback. Technologies: Python, TensorFlow, PyTorch, scikit-learn, EfficientNet, YOLO, Vision Transformer, and Edge Deployment.
Arxia – Agentic RAG Research Assistant
Built an agentic RAG chatbot enabling natural language conversations over 25 landmark AI/ML research papers from ArXiv, with source-attributed answers and page-level citations. Designed a 5-category intent classification system using zero-temperature Gemini to route messages - eliminating unnecessary vector. Implemented hybrid retrieval combining FAISS vector search (60%) and BM25 keyword search (40%) using LangChain EnsembleRetriever, improving recall for both semantic queries and exact terminology. Architected a two-layer memory system - short-term session memory and long-term cross-session memorY, injected into the system prompt for personalised responses. Deployed FastAPI backend on AWS EC2 with Ngrok HTTPS tunnel and React frontend on Vercel. Technologies: Python, FastAPI, LangChain, HuggingFace Transformers, FAISS, BM25, Google Gemini, SQLite, React, AWS EC2, Vercel.
AI Resume Analyzer
Designed and developed an end-to-end AI-powered resume analysis system using FastAPI, React, and Gemini 2.5 for ATS-style evaluation. Built a multi-stage pipeline with PDF parsing, keyword extraction, semantic similarity (all-MiniLM-L6-v2), and LLM-based structured analysis. Implemented a scoring engine combining keyword match, semantic alignment, and skills coverage to generate interpretable scores and grades. Developed skill gap detection and recommendation system, providing actionable feedback on missing technologies and improvements. Technologies: Python, FastAPI, React, Tailwind, Gemini 2.5 Flash, and Sentence-Transformers.
CERTIFICATION
Machine Learning Specialization
Comprehensive training in ML algorithms and implementations
Deep Learning: Neural Networks, RNNs, LSTM, GRU
Advanced neural network architectures
Natural Language Processing & Large Language Models
Modern NLP and LLM techniques
AI Agents Fundamentals
Building intelligent autonomous agents