<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
  <title>viot · lab notes</title>
  <link href="https://viotnet.pages.dev/"/>
  <link rel="self" href="https://viotnet.pages.dev/feed.xml"/>
  <id>https://viotnet.pages.dev/</id>
  <updated>2026-10-16T00:00:00Z</updated>
  <author><name>Konstantinos Papakonstantinou</name></author>
  <entry>
    <title>A Windows 11 forensics case, and an AI agent for triage</title>
    <link href="https://viotnet.pages.dev/blog/windows-11-forensics-ai-agent/"/>
    <id>https://viotnet.pages.dev/blog/windows-11-forensics-ai-agent/</id>
    <updated>2026-10-16T00:00:00Z</updated>
    <summary>A course forensics case on an encrypted Windows 11 VM: handling the evidence, mapping the findings to MITRE ATT&amp;CK, and what an LLM agent can and cannot do for triage.</summary>
    <category term="forensics"/><category term="dfir"/><category term="windows"/><category term="llm"/>
  </entry>
  <entry>
    <title>Designing a secure smart-city IoT architecture</title>
    <link href="https://viotnet.pages.dev/blog/secure-smart-city-architecture/"/>
    <id>https://viotnet.pages.dev/blog/secure-smart-city-architecture/</id>
    <updated>2026-10-16T00:00:00Z</updated>
    <summary>A layered edge-to-cloud design for city cameras, traffic sensors and street lighting, and the controls that keep one hacked lamp post from reaching the rest.</summary>
    <category term="architecture"/><category term="iot"/><category term="privacy"/>
  </entry>
  <entry>
    <title>Madrid air quality: clustering, classification and transfer learning</title>
    <link href="https://viotnet.pages.dev/blog/madrid-air-quality-machine-learning/"/>
    <id>https://viotnet.pages.dev/blog/madrid-air-quality-machine-learning/</id>
    <updated>2026-10-16T00:00:00Z</updated>
    <summary>A team project on Madrid air-quality data: K-Means against DBSCAN, SVM against Naive Bayes, and a small neural network that behaved better with transfer learning.</summary>
    <category term="machine learning"/><category term="data science"/><category term="team project"/>
  </entry>
  <entry>
    <title>A network log analysis agent that keeps the evidence in-house</title>
    <link href="https://viotnet.pages.dev/blog/local-llm-network-log-analysis/"/>
    <id>https://viotnet.pages.dev/blog/local-llm-network-log-analysis/</id>
    <updated>2026-10-16T00:00:00Z</updated>
    <summary>Detecting beaconing, DNS tunnelling, lateral movement and exfiltration in network logs with plain statistics, then letting a small local LLM write the summary.</summary>
    <category term="forensics"/><category term="network"/><category term="llm"/>
  </entry>
  <entry>
    <title>A motion and comfort sensor app, with replay protection</title>
    <link href="https://viotnet.pages.dev/blog/iot-activity-sensor-replay-protection/"/>
    <id>https://viotnet.pages.dev/blog/iot-activity-sensor-replay-protection/</id>
    <updated>2026-10-16T00:00:00Z</updated>
    <summary>A team project for an IoT security course: recognising posture and falls from an accelerometer, a heat-index comfort reading, and a nonce against replayed messages.</summary>
    <category term="iot"/><category term="sensors"/><category term="team project"/>
  </entry>
  <entry>
    <title>I audited my own password manager, then rewrote it</title>
    <link href="https://viotnet.pages.dev/blog/rewriting-my-password-manager/"/>
    <id>https://viotnet.pages.dev/blog/rewriting-my-password-manager/</id>
    <updated>2026-10-15T00:00:00Z</updated>
    <summary>Good primitives are not a secure system. What a security audit of my own Python password manager found, how version 2 is designed, and five bugs that only appeared when I ran it.</summary>
    <category term="write-up"/><category term="cryptography"/><category term="python"/>
  </entry>
</feed>
