<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Michał Michalski</title><description>CTO &amp; Lead AI Software Engineer. Thoughts on software engineering, AI agents, and things I find interesting.</description><link>https://buyuk.io/</link><item><title>Streaming EEG Data from Muse Direct – A Beginner’s Guide</title><link>https://buyuk.io/blog/muse2-eeg-osc-streaming/</link><guid isPermaLink="true">https://buyuk.io/blog/muse2-eeg-osc-streaming/</guid><description>Exploring Brain-Computer Interfaces (BCI) led me to experiment with consumer-grade EEG devices, ultimately choosing the Muse 2 for its signal quality. However, I soon discovered that Interaxon no longer supports the official API. Fortunately, streaming EEG data is still possible using the OpenSound Control (OSC) protocol and Muse Direct. In this post, I walk through setting up Muse Direct, configuring Muse Lab to visualize EEG signals, and verifying the data stream. Next time, I’ll dive into coding a Python script to process EEG data for BCI applications like mind-controlled Pong!</description><pubDate>Tue, 24 Mar 2020 00:00:00 GMT</pubDate></item><item><title>Measuring Frequency Response on a Budget – Using a Computer as a Signal Generator</title><link>https://buyuk.io/blog/sound-card-signal-generator/</link><guid isPermaLink="true">https://buyuk.io/blog/sound-card-signal-generator/</guid><description>When designing electronic circuits, testing the frequency response of components can be crucial for debugging and performance tuning. While professional spectrum analyzers and signal generators can be expensive, a simple alternative exists—using a computer’s sound card and an old pair of headphones. In this post, I demonstrate how to generate a sine wave signal using YouTube videos and measure it with an oscilloscope. This budget-friendly approach is perfect for electronics beginners. Stay tuned for the next post, where I’ll explore creating a custom sine wave generator in Python!</description><pubDate>Tue, 05 Nov 2019 00:00:00 GMT</pubDate></item><item><title>Tracing LangChain Embeddings with Langfuse</title><link>https://buyuk.io/blog/tracing-langchain-embeddings-with-langfuse/</link><guid isPermaLink="true">https://buyuk.io/blog/tracing-langchain-embeddings-with-langfuse/</guid><description>LangChain&apos;s embedding models don&apos;t fire callbacks like chat models do, leaving your embedding operations invisible in Langfuse traces. Here&apos;s a simple wrapper that adds full observability to any LangChain embeddings model.</description><pubDate>Wed, 26 Nov 2025 00:00:00 GMT</pubDate></item><item><title>ECG with Arduino and SparkFun AD8232 – Capturing Bioelectric Signals</title><link>https://buyuk.io/blog/ecg-with-arduino-and-sparkfun-ad8232-sensor/</link><guid isPermaLink="true">https://buyuk.io/blog/ecg-with-arduino-and-sparkfun-ad8232-sensor/</guid><description>Exploring bioelectric signals led me to experiment with ECG sensors as a stepping stone toward building a custom EEG circuit. In this post, I demonstrate how to connect the SparkFun AD8232 ECG sensor to an Arduino UNO, read the signal via a serial port, and visualize real-time ECG data using Python (matplotlib). This simple setup provides valuable insights into bioelectric monitoring. Next, I plan to experiment with a wireless ECG sensor and even build a DIY version. Full source code is available on GitHub!</description><pubDate>Thu, 21 Nov 2019 00:00:00 GMT</pubDate></item><item><title>Introducing FMQL: a query language for markdown frontmatter workspaces</title><link>https://buyuk.io/blog/fmql-launch/</link><guid isPermaLink="true">https://buyuk.io/blog/fmql-launch/</guid><description>fmql treats any directory of markdown files with YAML frontmatter as a schemaless document database: typed filters, reference traversal, bulk edits, pluggable search, and an agile-board example.</description><pubDate>Fri, 17 Apr 2026 00:00:00 GMT</pubDate></item><item><title>My LinkedIn Network Breakdown</title><link>https://buyuk.io/blog/whats-in-my-linkedin-network/</link><guid isPermaLink="true">https://buyuk.io/blog/whats-in-my-linkedin-network/</guid><description>I exported my 1k+ LinkedIn connections and analyzed them using AI, then rendered everything on a dashboard. You can do that too.</description><pubDate>Thu, 23 Apr 2026 00:00:00 GMT</pubDate></item></channel></rss>