<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>罗一阳's Insights</title><link>https://yiyang92.github.io/</link><description>Recent content on 罗一阳's Insights</description><image><title>罗一阳's Insights</title><url>https://yiyang92.github.io/%3Clink%20or%20path%20of%20image%20for%20opengraph,%20twitter-cards%3E</url><link>https://yiyang92.github.io/%3Clink%20or%20path%20of%20image%20for%20opengraph,%20twitter-cards%3E</link></image><generator>Hugo -- 0.146.0</generator><language>en</language><lastBuildDate>Sat, 14 Mar 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://yiyang92.github.io/index.xml" rel="self" type="application/rss+xml"/><item><title>Paper Clusterer: An Obsidian Plugin for Research Note Organization</title><link>https://yiyang92.github.io/posts/pet-projects/paper-clusterer-obsidian-plugin/</link><pubDate>Sat, 14 Mar 2026 00:00:00 +0000</pubDate><guid>https://yiyang92.github.io/posts/pet-projects/paper-clusterer-obsidian-plugin/</guid><description>&lt;h1 id="paper-clusterer-turn-your-obsidian-vault-into-a-research-map">Paper Clusterer: Turn Your Obsidian Vault into a Research Map&lt;/h1>
&lt;p>&lt;img alt="Paper Clusterer Visualization - Cluster Overview" loading="lazy" src="https://yiyang92.github.io/images/image-101.png">&lt;/p>
&lt;p>&lt;img alt="Paper Clusterer Visualization - Focused Cluster" loading="lazy" src="https://yiyang92.github.io/images/image-102.png">&lt;/p>
&lt;h2 id="the-problem-a-graveyard-of-research-notes">The Problem: A Graveyard of Research Notes&lt;/h2>
&lt;p>My Obsidian vault had become a mess. Hundreds of paper notes, lecture materials, and research ideas—scattered across folders, poorly connected, and increasingly difficult to navigate. Sound familiar?&lt;/p>
&lt;p>As researchers, we consume vast amounts of information. We take notes, highlight papers, jot down ideas. But without proper organization, this knowledge remains fragmented and underutilized.&lt;/p></description></item><item><title>From Sound to Meaning: Leveraging Audio Language Models for Music Relevance Assessment</title><link>https://yiyang92.github.io/posts/audio-llm/from-sound-to-meaning-audio-llm-music-relevance/</link><pubDate>Sat, 07 Mar 2026 00:00:00 +0000</pubDate><guid>https://yiyang92.github.io/posts/audio-llm/from-sound-to-meaning-audio-llm-music-relevance/</guid><description>How foundation audio models and Audio LLMs bridge the gap between acoustic similarity and perceptual relevance</description></item><item><title>[AI for all] What is Intelligence?</title><link>https://yiyang92.github.io/posts/llm/what-is-intelligence-ai-for-all/</link><pubDate>Wed, 29 Jan 2025 00:00:00 +0000</pubDate><guid>https://yiyang92.github.io/posts/llm/what-is-intelligence-ai-for-all/</guid><description>&lt;h1 id="ai-for-all-what-is-intelligence-">[AI for all] What is Intelligence? 🤔&lt;/h1>
&lt;p>So what exactly is intelligence? Intelligence is what makes us &amp;ldquo;Homo sapiens&amp;rdquo; - literally wise humans, thinking beings. What makes us intellectual is the fact that we can expect that the result of our actions will lead us to achieve our goals.&lt;/p>
&lt;p>But what makes machines intelligent? The same thing - their actions should lead to achieving their goals. The question is that we set the goals, and machines learn to optimize the process of achieving these goals in the best way. And we must be absolutely sure that we set the goals correctly, otherwise we can over-optimize, especially if (when?) we invent machines that will be much smarter than us and will quite be able to optimize us too :)&lt;/p></description></item><item><title>Beijing Car Visa: The Capital's Invisible Border</title><link>https://yiyang92.github.io/posts/china/beijing-visa-car-permit/</link><pubDate>Tue, 28 Jan 2025 00:00:00 +0000</pubDate><guid>https://yiyang92.github.io/posts/china/beijing-visa-car-permit/</guid><description>&lt;h1 id="beijing-car-visa-the-capitals-invisible-border">Beijing Car Visa: The Capital&amp;rsquo;s Invisible Border&lt;/h1>
&lt;p>The role of capital cities is crucial in many countries, but in nations with governance structures similar to China—where the Party plays a leading role and the political system closely resembles the Soviet model—the capital&amp;rsquo;s role becomes particularly significant.&lt;/p>
&lt;p>While Russians joke about the &amp;ldquo;rubber city&amp;rdquo; of Moscow and its hyper-centralization, moving to or even driving to Moscow isn&amp;rsquo;t particularly problematic. You can drive into Moscow with regional license plates without major issues.&lt;/p></description></item><item><title>Education in China: From Disappointment to Opportunity</title><link>https://yiyang92.github.io/posts/china/education-disappointments-opportunities/</link><pubDate>Mon, 27 Jan 2025 00:00:00 +0000</pubDate><guid>https://yiyang92.github.io/posts/china/education-disappointments-opportunities/</guid><description>&lt;h1 id="education-in-china-from-disappointment-to-opportunity">Education in China: From Disappointment to Opportunity&lt;/h1>
&lt;p>&lt;em>This might be a bit rambling, but here we go.&lt;/em>&lt;/p>
&lt;h2 id="the-initial-dream">The Initial Dream&lt;/h2>
&lt;p>My biggest disappointment at that time was probably my university experience. One of my main motivations for coming to China to study was to continue developing in mechanical engineering—to see how Chinese engineers work and how factories operate.&lt;/p>
&lt;p>&lt;strong>Reality check:&lt;/strong> I didn&amp;rsquo;t set foot in a Chinese factory until 2024&amp;hellip; but I&amp;rsquo;m getting ahead of myself.&lt;/p></description></item><item><title>Thoughts on Learning Chinese</title><link>https://yiyang92.github.io/posts/china/thoughts-on-chinese-language/</link><pubDate>Sun, 26 Jan 2025 00:00:00 +0000</pubDate><guid>https://yiyang92.github.io/posts/china/thoughts-on-chinese-language/</guid><description>&lt;h1 id="thoughts-on-learning-chinese">Thoughts on Learning Chinese&lt;/h1>
&lt;p>Yes, I speak, write, and understand Chinese. On one hand—it sounds impressive. On the other—why bother, really?&lt;/p>
&lt;h2 id="the-practical-truth">The Practical Truth&lt;/h2>
&lt;p>For the most part, Chinese turned out to be useful more for personal development (and for talking with my wife—obviously :)). Is it frustrating that the language is barely needed at work? I think so. Though when it was needed, Chinese corporate culture always came as a bonus&amp;hellip; I should probably write about that separately sometime.&lt;/p></description></item><item><title>About &amp; Contact</title><link>https://yiyang92.github.io/about/</link><pubDate>Sun, 28 Jan 2024 00:00:00 +0000</pubDate><guid>https://yiyang92.github.io/about/</guid><description>&lt;div class="about-content">
&lt;div class="about-hero">
&lt;!-- Photo placeholder - add your photo here -->
&lt;div class="about-hero-media">
&lt;img src="https://yiyang92.github.io/images/profile-photo.jpg" alt="Nikolai Zakharov" class="about-photo">
&lt;/div>
&lt;div class="about-hero-text">
&lt;h1 id="nikolai-zakharov-罗一阳">Nikolai Zakharov (罗一阳)&lt;/h1>
&lt;p>&lt;strong>ML Engineer at &lt;a href="https://vk.company/en/">VK AI&lt;/a>&lt;/strong>, working with the Experimental Research Team on LLM + audio systems (since Jan 2026).&lt;/p>
&lt;p>I build language models that understand audio and speech, bridging the gap between text and audio modalities. Based in Moscow after 10 years living in China, studying and working across telecom, fintech, and EVs (Huawei, Tinkoff, NIO). Prior to VK AI, I worked on recommender systems at VK Video (2025).&lt;/p></description></item><item><title>CV / Resume</title><link>https://yiyang92.github.io/cv/</link><pubDate>Sun, 28 Jan 2024 00:00:00 +0000</pubDate><guid>https://yiyang92.github.io/cv/</guid><description>&lt;h1 id="nikolai-zakharov-罗一阳">Nikolai Zakharov (罗一阳)&lt;/h1>
&lt;p>&lt;strong>Senior Machine Learning Engineer (Multimodal / LLM)&lt;/strong>&lt;/p>
&lt;p>Moscow, Russia (Open to China/Singapore/Global)&lt;br>
Tel: +86 136 2105 3870&lt;br>
Email: &lt;a href="mailto:n.d.zakharov@outlook.com">n.d.zakharov@outlook.com&lt;/a>&lt;br>
LinkedIn: &lt;a href="https://www.linkedin.com/in/nikolai-zakharov-285841a7/">https://www.linkedin.com/in/nikolai-zakharov-285841a7/&lt;/a>&lt;br>
Website: &lt;a href="https://yiyang92.github.io/">https://yiyang92.github.io/&lt;/a>&lt;/p>
&lt;p>&lt;a href="https://yiyang92.github.io/files/cv.pdf">Download English CV (PDF) →&lt;/a>  |  &lt;a href="https://yiyang92.github.io/files/cv-cn.pdf">下载中文简历 (PDF) →&lt;/a>&lt;/p>
&lt;hr>
&lt;h2 id="summary">Summary&lt;/h2>
&lt;p>Senior Machine Learning Engineer (Multimodal / LLM). 8+ years production AI at scale. Specialized in &lt;strong>Audio LLMs&lt;/strong>, &lt;strong>multimodal systems&lt;/strong>, &lt;strong>inference optimization&lt;/strong> (vLLM, TensorRT). Delivered: captioning service for music recommendations (VK), RAG agents (+20%, NIO), voice synthesis (-88% latency, Tinkoff). Tsinghua MSc, 10 years China tech ecosystem. Fluent in Russian, English, Chinese.&lt;/p></description></item></channel></rss>