TECHNICAL EXPERTISE & ENGINEERING STACK

Asma Rizwan Rao is an independent software developer focused on the architecture and development of AI-driven systems, neurotechnology software, intelligent automation, and computer-vision applications. Her work integrates machine learning, signal-processing pipelines, agentic architectures, API-driven applications, and real-time visualization into cohesive software systems.

01 — Artificial Intelligence & Machine Learning

Core Technologies: PyTorch, scikit-learn, NumPy, SciPy, pandas, Matplotlib, Ultralytics YOLOv8.

Model Architectures & Methods: Convolutional Neural Networks (CNNs), EEGNet, Logistic Regression, Linear Discriminant Analysis (LDA), Common Spatial Patterns (CSP), supervised classification, feature engineering, model inference, and computer-vision pipelines.

Engineering Focus: Dataset preparation, preprocessing, model integration, inference workflows, performance evaluation, and deployment of machine-learning models within interactive applications.

02 — Neurotechnology Software & Neural Signal Processing

Technical Stack: MNE-Python, Python scientific-computing libraries, EDF-compatible EEG data workflows, PhysioNet datasets.

Signal Processing: Digital filtering, 50 Hz power-line notch filtering, 8–30 Hz bandpass filtering, channel-wise z-score normalization, EEG epoch preparation, motor-imagery classification, and event-related desynchronization/synchronization (ERD/ERS).

BCI Modeling & Evaluation: EEGNet-based classification, Logistic Regression baselines, three-class and five-class motor-imagery decoding, subject-specific calibration, cross-subject generalization analysis, and Leave-One-Subject-Out (LOSO) evaluation where implemented.

Engineering Focus: End-to-end EEG data pipelines spanning signal preprocessing, feature representation, model inference, classification outputs, and user-facing visualization.

03 — Agentic AI & Intelligent Automation

Frameworks & Protocols: LangChain, LangGraph, n8n, Model Context Protocol (MCP).

System Capabilities: Tool-using AI agents, multi-step workflow orchestration, conversational interfaces, memory-enabled workflows, Retrieval-Augmented Generation (RAG), structured output generation, API-connected tools, and prompt design.

Engineering Focus: Building application-oriented AI systems that connect language models, external services, business workflows, and software interfaces into coordinated processes.

04 — Software Architecture & Application Engineering

Programming Languages: Python, C, SQL, JavaScript, HTML5, CSS3.

Application Frameworks: FastAPI, Streamlit, Tkinter.

Engineering Disciplines: Modular application architecture, REST API integration, data processing, asynchronous and event-driven workflows where implemented, state management, input validation, exception handling, and interface development.

Business Systems: CRM application architecture, sales pipeline automation, workflow orchestration, SaaS application design, and integration of AI capabilities into business software.

05 — Computer Vision, Tracking & Geospatial Visualization

Technologies: OpenCV, Ultralytics YOLOv8, Matplotlib, TkinterMapView, GeoPy.

Technical Areas: Image acquisition, object detection, bounding-box visualization, confidence-score interpretation, coordinate tracking, time-series telemetry, rolling data visualization, and geographic mapping.

Engineering Focus: Integrating visual inference, application interfaces, tracking data, and geospatial representations into a unified desktop software environment.

06 — Electromagnetic Effects Modeling & Simulation Software

Technical Areas: Numerical modeling, parameterized physical simulations, signal and waveform representation, attenuation modeling, and software-based effects visualization.

Engineering Focus: Implementing computational models as modular software components, separating model assumptions from calculated outputs, and documenting numerical methods, limitations, and validation requirements.

Project Context: SkyNet (A818_SK) includes a dedicated electromagnetic-effects simulation module. Its simulated outputs should be distinguished from experimentally validated physical effects.

07 — APIs, Integration & Intelligent Business Systems

Technologies: FastAPI, REST APIs, SQL, Python automation libraries, LangChain, LangGraph, n8n, MCP.

Technical Areas: Service integration, request/response handling, structured data exchange, workflow automation, external-tool integration, business-process orchestration, and modular application design.

Engineering Focus: Connecting independent services and computational components through maintainable interfaces to build functional, extensible software systems.

08 — Development Infrastructure & Deployment

Version Control: Git, GitHub, GitHub Desktop.

Compute Environments: Google Colab, NVIDIA T4 GPU environments, Kaggle.

Application Hosting & Tunneling: Streamlit Community Cloud, GitHub Pages, Cloudflare Tunnel.

Development Practices: Dependency management, environment configuration, repository organization, reproducible setup instructions, debugging, deployment troubleshooting, and version-controlled development.

09 — Datasets & Research Resources

EEG & BCI: PhysioNet EEG Motor Movement/Imagery dataset; BCI Competition IV Dataset 2a where used.

Computer Vision: Seraphim drone-detection dataset, as identified in the SkyNet project documentation.

Research Workflow: Dataset inspection, preprocessing, experiment configuration, model evaluation, and documentation of data provenance and experimental limitations.

SELECTED INDEPENDENT SOFTWARE SYSTEMS

NeuroFlow — A818_NF
An EEG motor-imagery classification application integrating signal preprocessing, machine-learning inference, and an interactive Streamlit interface. The project reports three-class Logistic Regression performance of approximately 90.78% and a five-class EEGNet result of approximately 78% under its documented evaluation setups.

SkyNet — A818_SK
A Python desktop application combining YOLO-based aerial-object detection, visual telemetry, geographic visualization, and an electromagnetic-effects simulation module. Hardware-dependent functionality remains distinct from the software components and requires separate validation.

Agentic AI & Automation Projects
Application-oriented systems exploring AI agents, tool integration, workflow orchestration, and automated business processes.

Engineering Philosophy: Build integrated systems, make their