Problem: Need for a lightweight, browser-based, and code-accessible tool for music production.

Solution: Bundler-free architecture with vanilla JS + ES modules, built on Strudel live-coding engine, running from a single index.html.

Result: Single-page mini DAW application with pattern-based drum machine, sequencer, and code editor.

Problem: Corporate firms need for fast, modern, and conversion-focused landing pages.

Solution: Zero-JS performance with Astro and dynamic content with Next.js, responsive and SEO-optimized designs.

Result: High-performance landing page solutions aligned with corporate identity.

Problem: Database querying and reporting processes requiring technical knowledge.

Solution: Natural language database querying and automatic report generation.

Result: Self-service reporting capability for non-technical users.

Problem: Limited solutions for contactless and real-time pulse measurement.

Solution: Integration of face recognition and pulse detection models from literature with C++, real-time video processing.

Result: Real-time pulse value display via camera.

Problem: Difficulty in manual plant watering and moisture/temperature tracking.

Solution: Sensor-based automatic irrigation and data monitoring system.

Result: Optimized water usage and plant health tracking.

Problem: Need to compare classical and deep learning models for stock price prediction.

Solution: Data exploration with EDA, statistical prediction with ARIMA, deep learning-based prediction with LSTM, comparative reporting.

Result: Comparative analysis and visual reports of both models' performance metrics.

Problem: Need to transform raw data into meaningful insights.

Solution: Analytical reporting with statistical modeling and visualization.

Result: Data-driven decision support systems.

Problem: Manual and fragmented customer acquisition and identity verification processes.

Solution: Integrated architectural design of web-based KYC, video call, and monitoring systems.

Result: End-to-end automated customer verification process.

Problem: Contact-requiring methods and security vulnerabilities in restaurant ordering processes.

Solution: Contactless and secure flow with QR-based ordering, crypto payment, and face recognition.

Result: Contactless ordering and payment experience with enhanced security.

Problem: Compatibility challenges in deploying AI models across different platforms.

Solution: Cross-platform model deployment through ONNX format conversion.

Result: Optimized model performance in mobile and web environments.