Mahmut Sarikaya - Software Developer
Kategoriler
Blog
Deploy .NET 8 Native AOT to AWS Lambda for Ultra‑Fast Serverless Functions
Explore how to compile .NET 8 projects with Native AOT into lightweight native binaries and run them as custom runtimes on AWS Lambda. The guide walks through project setup, dependency trimming, packaging, and deployment, showing measurable latency improvements and cost benefits for serverless C# workloads.
Build Real-Time Collaborative Mapping with .NET 8, SignalR, and Azure Maps
Learn how to build interactive, multi‑user mapping solutions using .NET 8 Minimal APIs, SignalR for low‑latency WebSocket communication, and Azure Maps for geospatial rendering. The guide covers architecture design, authentication with Azure AD, scaling strategies, and practical code samples for synchronizing map data across browsers in real time.
Designing Event-Driven Serverless Architectures with .NET 8, Azure Functions, and Dapr
Explore how to combine .NET 8's latest performance features with Azure Functions and Dapr to build scalable, event‑driven serverless applications. The guide covers setting up Dapr sidecars, publishing/subscribing to CloudEvents, and deploying C# function apps that react instantly to changes in storage, messaging, or APIs.
Real-Time AI-Powered Content Moderation in the Browser with TensorFlow.js, Web Workers, and WebGPU
Discover how to build an on‑device, privacy‑preserving content moderation pipeline that runs entirely in the browser. This tutorial walks through selecting lightweight TensorFlow.js models, accelerating inference with WebGPU, offloading computation to Web Workers for smooth UI, and integrating the system into chat or comment sections for instant abuse detection.
Run Transformer Text Summarization in the Browser with WebGPU & TensorFlow.js
Learn how to leverage WebGPU acceleration and TensorFlow.js to load and run lightweight transformer models entirely in the browser for instant text summarization. The guide covers model conversion, performance tuning with GPU shaders, and practical integration into JavaScript web apps.
Building AI‑Powered Real‑Time Code Assistants in the Browser
Learn how to create a responsive, in‑browser code assistant that leverages OpenAI's function‑calling API, runs heavy parsing and suggestion logic in WebAssembly, and integrates seamlessly with modern JavaScript editors. The guide covers setting up the OpenAI SDK, compiling a lightweight analysis engine to WASM, and wiring real‑time feedback into VS Code extensions or web‑based IDEs.
Build Serverless Edge AI Functions with Deno Deploy and Supabase Edge Runtime
Learn how to create ultra‑low latency AI endpoints that run at the network edge using Deno Deploy and Supabase Edge Runtime. The guide covers deploying WebAssembly‑based models, integrating Supabase's edge database, and exposing secure serverless JavaScript functions for real‑time inference.
Build AI-Powered Minimal APIs with .NET 8, Azure OpenAI & Semantic Kernel
Learn how to create lightweight, high‑performance Minimal API services in .NET 8 that harness Azure OpenAI and Microsoft Semantic Kernel to add natural‑language understanding, content generation, and reasoning capabilities. The guide covers project scaffolding, authentication with Azure AD, prompt engineering, async streaming responses, and deployment to Azure Container Apps with best‑practice observability and scaling.
Harnessing WebNN API for Hardware‑Accelerated AI in JavaScript
Explore the emerging WebNN API that lets JavaScript code run neural networks directly on device GPUs, NPUs, and other accelerators. Learn how to set up a WebNN pipeline, integrate it with WebGPU for custom kernels, and build real‑world use cases such as on‑device image classification and speech recognition without server round‑trips.
Edge AI Inference with ONNX Runtime Web & WebGPU on Cloudflare Workers
Learn how to run high‑performance, quantized AI models directly at the edge using ONNX Runtime Web accelerated by WebGPU. The guide covers setting up Cloudflare Workers, loading ONNX models, optimizing them for WebGPU, and integrating the inference pipeline into a JavaScript front‑end for ultra‑low latency predictions.
Running Stable Diffusion in Browser with JavaScript & WebGPU
Learn how to load, optimize, and execute the Stable Diffusion model directly in the browser using JavaScript and the WebGPU API. The tutorial covers model conversion to ONNX, leveraging ONNX Runtime Web or TensorFlow.js, GPU‑accelerated inference, and real‑time UI integration for on‑device image generation without server dependencies.
Client‑Side Scientific Simulations with WebGPU Compute Shaders in JavaScript
Explore how to run high‑performance scientific simulations directly in the browser by leveraging WebGPU's compute pipeline with JavaScript. The guide walks through environment setup, writing and dispatching compute shaders, integrating numeric libraries, and real‑world use cases such as fluid dynamics, particle systems, and cellular automata.