I'm a passionate technologist and the CEO of AppCraft Studios, focused on architecting intelligent software that solves complex problems.
- Artificial Intelligence & Machine Learning: Designing multi-model AI pipelines, particularly in computer vision, OCR, and predictive analytics.
- Full-Stack & Systems Architecture: Building robust, scalable platforms spanning enterprise ASP.NET Core MVC backends, modern React/Next.js interfaces, low-level runtime engines, and off-grid mesh protocols.
- Mobile Application Development: Crafting intuitive, high-performance mobile experiences using Flutter, Kotlin, and Jetpack Compose.
- Data Science & Financial Intelligence: Extracting actionable insights, reconciling unstructured transactional documents, and engineering offline geospatial mapping systems.
- Beyond Tech: Keenly interested in astronomy, aviation, and aerial photography.
- Languages: Python, Dart, C#, TypeScript, JavaScript, Kotlin, Java, C++, SQL, PHP, Delphi, HTML, CSS.
- Frameworks & Libraries: Flutter, React, Next.js, ASP.NET Core MVC, Jetpack Compose, Node.js, Flask, Blazor, TensorFlow, PyTorch, Scikit-Learn, NumPy, MapLibre.
- AI & Vision: Computer Vision, OCR Pipelines, Document Parsing, RAG, Anomaly Detection, CNNs, Generative AI SDKs, OpenCV.
- Cloud & DevOps: Google Cloud Platform (GCP), Microsoft Azure, Firebase, Supabase, Vercel, RevenueCat.
- Databases: PostgreSQL, SQLite (Drift), Room (Android), MongoDB, SQL Server.
- Hardware & Systems: WebAssembly (Wasm), LoRa RF Mesh, BLE GATT Firmware (FreeRTOS), Android CameraX, RESTful APIs, Role-Based Access Control (RBAC), Agile Methodologies.
- FinYang AI π: Universal bank statement forensic parsing and underwriting engine that reconciles multi-account financial ledgers to the cent with automated credit memorandums - https://finyang.net/
- CapeFin AI πΌ: Enterprise credit underwriting and financial document extraction platform engineered for asset-based finance and logistics risk assessment.
- AfriBush πΎ: An off-grid African wilderness OS and LoRa/BLE mesh platform featuring sub-30 MB offline satellite GIS, deterministic FGASA trait matrices, and anti-poaching safeguards - https://afribush.com/
- Slip-Scan π§Ύ: Commercial AI-driven receipt and invoice management platform built with Flutter, featuring batch OCR scanning, offline synchronisation, and direct integrations with Xero and Sage - https://www.slip-scan.com/
- Skye Elevators Operations Platform π: Dual-interface management and field operations suite built with Flutter and React, implementing role-based access control (RBAC), live technician allocation, and asset service logging.
- Universal Agentic Framework β‘: Experimental neural runtime engine designed to compile natural language commands directly into WebAssembly binary bytecode for native execution speeds - https://jamie123234.github.io/universal_agentic_framework/
- MoneyQuest πͺ: Gamified personal budget tracking Android application engineered with Jetpack Compose, local Room database persistence, and Firebase cloud synchronisation.
- RecruitAI π: Intelligent recruitment platform using backend scrapers and candidate matching algorithms for automated CV evaluation and unbiased job alignment - https://recruitai.co.za/
- Project Aether's Eye π½ (Coming Soon): Autonomous astronomical anomaly detection pipeline with a dual-AI core, combining a TensorFlow CNN for classification and a PyTorch neural network for event prioritisation - https://jamie123234.github.io/project_aether/
- Botavia πΏ: Native Android application using CameraX and Jetpack Compose for real-time camera previewing and optical nature specimen identification.
- Brain Tumor Diagnostic Model π§ : High-precision diagnostic classifier (>95% accuracy) engineered using NumPy and Scikit-Learn for structural brain anomaly identification from MRI scans.
- Global Logistics Management System π’: Enterprise logistics command centre developed with ASP.NET Core MVC, handling complex shipping schedules, multi-tier data processing, and relational database integrity.
- Royal Cape Yacht Club Systems Overhaul β΅: Led a comprehensive modernisation of core legacy operations and membership workflows, employing Flutter, Dart, and modern web technologies.
- Advanced techniques in ethical hacking, application security, and penetration testing.
- Compilers, WebAssembly runtimes, and low-level agentic execution architectures.
- Long-range off-grid radio mesh telemetry and resilient embedded field devices.
I'm always open to collaborating on challenging software engineering and AI projects, whether remote or locally in Cape Town. I am especially keen on opportunities in machine learning, full-stack enterprise systems, and mobile application engineering.
- Email: jamieabrahams123234@gmail.com
- LinkedIn: https://www.linkedin.com/in/jamie-abrahams/
- Website: https://jamieabrahams.com/

