Rabbiya Riaz

PROFESSIONAL PROFILE Software Engineer with hands-on experience building Python backend systems, ML-enabled applications, and automation workflows. Proven ability to debug and refactor production ML pipelines, improving model accuracy from ~10% to ~80% by correcting data preprocessing, resolving feature–model mismatches, and fixing system-level inconsistencies. Comfortable working across software engineering and applied AI, integrating ML models into reliable, scalable services rather than treating them in isolation. EDUCATION BS Computer Science SEECS , Islamabad , 3.14 (2026) INTERNSHIP EXPERIENCE School of Electrical Engineering and Computer Science (SEECS), NUST 03-Jul-2025 - 01-Sep-2025 Software / ML Intern: Debugged and modified core pipeline logic, including graph construction for skeleton data , to align system behavior with updated input specifications. Analyzed and restructured an existing Python-based ML automation pipeline to support a new data source, redesigning preprocessing workflows for compatibility and reliability. Improved pipeline accuracy from ~10% to ~80% by resolving data inconsistencies, fixing preprocessing failures, and validating end-to-end automation flow. Worked with an existing GitHub codebase to identify failure points and ensure correct execution across multiple datasets. FINAL YEAR PROJECT Federated Learning–Based Mental Health AI Platform (FYP) Applied federated learning principles to address data-privacy constraints in multi-client mental health systems, enabling privacy- preserving, client-isolated model training without centralized data sharing. Built a modular FastAPI backend to decouple authentication, AI logic, and data access, improving scalability, data isolation, and long-term maintainability. Integrated DistilBERT- based NLP pipelines and AI-driven REST APIs to support chat, questionnaires, and reporting workflows, enabling end-to-end ML feature integration. AI-Powered English Conversation Practice App (Academic Project) Used Speech-to-Text and large language model APIs (LLaMA) to enable real-time, context-aware conversational responses for spoken English practice. Implemented Text-to-Speech pipelines to provide natural, bidirectional voice feedback and improve user engagement. Built a cross-platform Flutter application integrated with Firebase (Auth, Firestore, Storage) to manage user state and persist conversation data across sessions. Amazon Price Tracker (Academic Project) Developed a hybrid web scraping system using Requests for lightweight pages and Selenium for dynamically rendered product pages to reliably track price changes. Implemented structured data extraction for price, availability, and ratings using HTML parsing with CSS/XPath selectors. Added automation features, including scheduled price checks and alert notifications when prices dropped below defined thresholds. TECHNICAL EXPERTISE Backend & APIs FastAPI, Flask, RESTful API Design, Authentication (OAuth 2.0, Firebase Auth)

Software Engineer with hands-on experience building Python backend systems, ML-enabled applications, and automation workflows. Proven ability to debug and refactor production ML pipelines, improving model accuracy from ~10% to ~80% by correcting data preprocessing, resolving feature–model mismatches, and fixing system-level inconsistencies. Comfortable working across software engineering and applied AI, integrating ML models into reliable, scalable services rather than treating them in isolation.