<?xml version="1.0" encoding="UTF-8"?>
<?xml-stylesheet href="/feed.rss.xml" type="text/xsl" media="screen"?>
<rss version="2.0" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:media="http://search.yahoo.com/mrss/" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Kannan Murugapandian</title>
    <description></description>
    <link>https://speakerdeck.com/kannanmp</link>
    <atom:link rel="self" type="application/rss+xml" href="https://speakerdeck.com/kannanmp.rss"/>
    <lastBuildDate>2026-03-29 06:46:54 -0400</lastBuildDate>
    <item>
      <title>Web Exploitation II [Lect 8: Competitive Cybersecurity 101]</title>
      <description>Transitioning from the browser to the backend, this lecture tackles severe server-side vulnerabilities where untrusted user input interacts directly with the operating system and databases. Students learn the mechanics and syntax of Command Injection and SQL Injection (SQLi), practicing classic login bypasses and arbitrary command execution. The session also introduces Burp Suite as the ultimate web proxy for intercepting and modifying HTTP requests.</description>
      <media:content url="https://files.speakerdeck.com/presentations/ae26cf4c0de6454481f1ef284354cb54/preview_slide_0.jpg?40888994" type="image/jpeg" medium="image"/>
      <content:encoded>Transitioning from the browser to the backend, this lecture tackles severe server-side vulnerabilities where untrusted user input interacts directly with the operating system and databases. Students learn the mechanics and syntax of Command Injection and SQL Injection (SQLi), practicing classic login bypasses and arbitrary command execution. The session also introduces Burp Suite as the ultimate web proxy for intercepting and modifying HTTP requests.</content:encoded>
      <pubDate>Tue, 13 Oct 2026 00:00:00 -0400</pubDate>
      <link>https://speakerdeck.com/kannanmp/web-exploitation-ii-lect-8-competitive-cybersecurity-101</link>
      <guid>https://speakerdeck.com/kannanmp/web-exploitation-ii-lect-8-competitive-cybersecurity-101</guid>
    </item>
    <item>
      <title>Web Exploitation I (Client-Side Attacks) [Lect 7: Competitive Cybersecurity 101]</title>
      <description>Focusing on the principle that "the client is never to be trusted," this lecture explores the architecture of web applications and the vulnerabilities inherent in client-side execution. Students learn to weaponize browser Developer Tools (F12) to manipulate the live DOM, bypass HTML input restrictions, and hijack session state by editing plaintext cookies. The session concludes by demonstrating the inherent flaws of client-side JavaScript validation, teaching students how to read source code to extract hardcoded credentials.</description>
      <media:content url="https://files.speakerdeck.com/presentations/af4166e698b14187b18a469f739fb825/preview_slide_0.jpg?40888872" type="image/jpeg" medium="image"/>
      <content:encoded>Focusing on the principle that "the client is never to be trusted," this lecture explores the architecture of web applications and the vulnerabilities inherent in client-side execution. Students learn to weaponize browser Developer Tools (F12) to manipulate the live DOM, bypass HTML input restrictions, and hijack session state by editing plaintext cookies. The session concludes by demonstrating the inherent flaws of client-side JavaScript validation, teaching students how to read source code to extract hardcoded credentials.</content:encoded>
      <pubDate>Tue, 13 Oct 2026 00:00:00 -0400</pubDate>
      <link>https://speakerdeck.com/kannanmp/web-exploitation-i-client-side-attacks-lect-7-competitive-cybersecurity-101</link>
      <guid>https://speakerdeck.com/kannanmp/web-exploitation-i-client-side-attacks-lect-7-competitive-cybersecurity-101</guid>
    </item>
    <item>
      <title>Forensics II (Networking &amp; Disk Forensics) [Lect 6: Competitive Cybersecurity 101]</title>
      <description>This session expands the forensic toolkit into network traffic analysis and disk imaging, emphasizing the golden rule of always analyzing bit-for-bit copies rather than original evidence. Students master Wireshark to filter millions of packets, intercept data exfiltration, and extract hidden files directly from HTTP traffic. The lecture also debunks the "deleted file" myth by exploring unallocated space and teaching file carving techniques using command-line tools and the Autopsy GUI to recover wiped evidence.</description>
      <media:content url="https://files.speakerdeck.com/presentations/e6d1da656d1343a6bc266d9d07e3b493/preview_slide_0.jpg?40888856" type="image/jpeg" medium="image"/>
      <content:encoded>This session expands the forensic toolkit into network traffic analysis and disk imaging, emphasizing the golden rule of always analyzing bit-for-bit copies rather than original evidence. Students master Wireshark to filter millions of packets, intercept data exfiltration, and extract hidden files directly from HTTP traffic. The lecture also debunks the "deleted file" myth by exploring unallocated space and teaching file carving techniques using command-line tools and the Autopsy GUI to recover wiped evidence.</content:encoded>
      <pubDate>Tue, 13 Oct 2026 00:00:00 -0400</pubDate>
      <link>https://speakerdeck.com/kannanmp/forensics-ii-networking-and-disk-forensics-lect-6-competitive-cybersecurity-101</link>
      <guid>https://speakerdeck.com/kannanmp/forensics-ii-networking-and-disk-forensics-lect-6-competitive-cybersecurity-101</guid>
    </item>
    <item>
      <title>Forensics I (Files, Metadata &amp; Steganography) [Lect 5: Competitive Cybersecurity 101]</title>
      <description>Grounded in Locard’s Exchange Principle, this lecture introduces the digital forensic mindset required to recover and analyze hidden evidence without altering the original files. Students learn to bypass deceptive file extensions by verifying Magic Bytes and extracting hidden metadata using industry-standard tools like `exiftool` and hex editors. The session also covers the art of steganography, teaching students how to uncover data hidden in image color planes (LSB) and audio spectrograms, as well as how to extract embedded files using `strings` and `binwalk`.  </description>
      <media:content url="https://files.speakerdeck.com/presentations/389f427f7b0444879ce9c4e1a219cd1c/preview_slide_0.jpg?40888840" type="image/jpeg" medium="image"/>
      <content:encoded>Grounded in Locard’s Exchange Principle, this lecture introduces the digital forensic mindset required to recover and analyze hidden evidence without altering the original files. Students learn to bypass deceptive file extensions by verifying Magic Bytes and extracting hidden metadata using industry-standard tools like `exiftool` and hex editors. The session also covers the art of steganography, teaching students how to uncover data hidden in image color planes (LSB) and audio spectrograms, as well as how to extract embedded files using `strings` and `binwalk`.  </content:encoded>
      <pubDate>Tue, 13 Oct 2026 00:00:00 -0400</pubDate>
      <link>https://speakerdeck.com/kannanmp/forensics-i-files-metadata-and-steganography-lect-5-competitive-cybersecurity-101</link>
      <guid>https://speakerdeck.com/kannanmp/forensics-i-files-metadata-and-steganography-lect-5-competitive-cybersecurity-101</guid>
    </item>
    <item>
      <title>Cryptography II (Asymmetric &amp; Post-Quantum) [Lect 4: Competitive Cybersecurity 101]</title>
      <description>This lecture tackles the key distribution problem by introducing asymmetric cryptography, focusing on the mathematical foundations of RSA and the Diffie-Hellman key exchange. Students learn how digital signatures reverse RSA keys to prove identity rather than secrecy. The session also looks to the future, exploring the existential threat that Shor's Algorithm and quantum computing pose to current encryption standards, and introduces Post-Quantum Cryptography (PQC) like lattice-based cryptography. Practical CTF skills include using `gmpy2`, FactorDB, and `RsaCtfTool` to exploit weak RSA implementations. </description>
      <media:content url="https://files.speakerdeck.com/presentations/ec5bb75ef67e436286dff52145d89027/preview_slide_0.jpg?40888823" type="image/jpeg" medium="image"/>
      <content:encoded>This lecture tackles the key distribution problem by introducing asymmetric cryptography, focusing on the mathematical foundations of RSA and the Diffie-Hellman key exchange. Students learn how digital signatures reverse RSA keys to prove identity rather than secrecy. The session also looks to the future, exploring the existential threat that Shor's Algorithm and quantum computing pose to current encryption standards, and introduces Post-Quantum Cryptography (PQC) like lattice-based cryptography. Practical CTF skills include using `gmpy2`, FactorDB, and `RsaCtfTool` to exploit weak RSA implementations. </content:encoded>
      <pubDate>Tue, 13 Oct 2026 00:00:00 -0400</pubDate>
      <link>https://speakerdeck.com/kannanmp/cryptography-ii-asymmetric-and-post-quantum-lect-4-competitive-cybersecurity-101</link>
      <guid>https://speakerdeck.com/kannanmp/cryptography-ii-asymmetric-and-post-quantum-lect-4-competitive-cybersecurity-101</guid>
    </item>
    <item>
      <title>Cryptography I (Classical &amp; Symmetric) [Lect 3: Competitive Cybersecurity 101]</title>
      <description>This session demystifies the core goals of cryptography—confidentiality, integrity, and authentication—while clarifying the critical distinctions between encoding, encryption, and hashing. Students explore classical ciphers like Caesar and Vigenère, learning to break them using Python modulo arithmetic and automated solvers like CyberChef and dCode.fr. The lecture then transitions to modern symmetric encryption, explaining the mathematical foundation of the XOR operation and the Advanced Encryption Standard (AES) as applied in CTF environments.</description>
      <media:content url="https://files.speakerdeck.com/presentations/4a4ebedc6ec8439888d829a48f6b0a06/preview_slide_0.jpg?40888805" type="image/jpeg" medium="image"/>
      <content:encoded>This session demystifies the core goals of cryptography—confidentiality, integrity, and authentication—while clarifying the critical distinctions between encoding, encryption, and hashing. Students explore classical ciphers like Caesar and Vigenère, learning to break them using Python modulo arithmetic and automated solvers like CyberChef and dCode.fr. The lecture then transitions to modern symmetric encryption, explaining the mathematical foundation of the XOR operation and the Advanced Encryption Standard (AES) as applied in CTF environments.</content:encoded>
      <pubDate>Tue, 13 Oct 2026 00:00:00 -0400</pubDate>
      <link>https://speakerdeck.com/kannanmp/cryptography-i-classical-and-symmetric-lect-3-competitive-cybersecurity-101</link>
      <guid>https://speakerdeck.com/kannanmp/cryptography-i-classical-and-symmetric-lect-3-competitive-cybersecurity-101</guid>
    </item>
    <item>
      <title>Python for CTF Scripting [Lect 2: Competitive Cybersecurity 101]</title>
      <description>Moving beyond basic syntax, this session focuses on writing robust, automated scripts to solve CTF challenges efficiently. Students learn to leverage Python dictionaries for fast data mapping and functions for modular "decoder toolboxes." The lecture covers essential forensic and web exploitation techniques, including memory-efficient File I/O for parsing massive logs, bulletproof exception handling (`try/except`), command-line arguments via `sys`, and making HTTP requests using the `requests` library. The session culminates in a hands-on automated log parser challenge</description>
      <media:content url="https://files.speakerdeck.com/presentations/6dd8c9fa1e6843e4a366dbd7ec0af495/preview_slide_0.jpg?40888785" type="image/jpeg" medium="image"/>
      <content:encoded>Moving beyond basic syntax, this session focuses on writing robust, automated scripts to solve CTF challenges efficiently. Students learn to leverage Python dictionaries for fast data mapping and functions for modular "decoder toolboxes." The lecture covers essential forensic and web exploitation techniques, including memory-efficient File I/O for parsing massive logs, bulletproof exception handling (`try/except`), command-line arguments via `sys`, and making HTTP requests using the `requests` library. The session culminates in a hands-on automated log parser challenge</content:encoded>
      <pubDate>Tue, 13 Oct 2026 00:00:00 -0400</pubDate>
      <link>https://speakerdeck.com/kannanmp/python-for-ctf-scripting-lect-2-competitive-cybersecurity-101</link>
      <guid>https://speakerdeck.com/kannanmp/python-for-ctf-scripting-lect-2-competitive-cybersecurity-101</guid>
    </item>
    <item>
      <title>Python Foundations [Lect 1b: Competitive Cybersecurity 101]</title>
      <description>Python is the undisputed industry standard for cybersecurity professionals, offering rapid prototyping, rich library support, and readable syntax essential for automating security tasks. This presentation constitutes Session 1 of the "Competitive Cybersecurity 101" series, bridging the gap between fundamental programming concepts and their direct application in Capture The Flag (CTF) competitions.

Moving beyond generic programming tutorials, this deck focuses exclusively on the data structures and operations most frequently encountered in cybersecurity challenges. Participants are guided through string manipulation, ASCII/Unicode translation, and the critical distinction between encoding (Hex, Base64) and encryption. The session culminates in a practical mini-challenge, requiring students to apply control flow and file I/O operations to programmatically extract and decode hidden flags, establishing the automation mindset required for advanced competitive informatics.</description>
      <media:content url="https://files.speakerdeck.com/presentations/ff4de6093b9944709a269221ee7077e6/preview_slide_0.jpg?40093974" type="image/jpeg" medium="image"/>
      <content:encoded>Python is the undisputed industry standard for cybersecurity professionals, offering rapid prototyping, rich library support, and readable syntax essential for automating security tasks. This presentation constitutes Session 1 of the "Competitive Cybersecurity 101" series, bridging the gap between fundamental programming concepts and their direct application in Capture The Flag (CTF) competitions.

Moving beyond generic programming tutorials, this deck focuses exclusively on the data structures and operations most frequently encountered in cybersecurity challenges. Participants are guided through string manipulation, ASCII/Unicode translation, and the critical distinction between encoding (Hex, Base64) and encryption. The session culminates in a practical mini-challenge, requiring students to apply control flow and file I/O operations to programmatically extract and decode hidden flags, establishing the automation mindset required for advanced competitive informatics.</content:encoded>
      <pubDate>Fri, 24 Jul 2026 00:00:00 -0400</pubDate>
      <link>https://speakerdeck.com/kannanmp/python-foundations-lect-1b-competitive-cybersecurity-101</link>
      <guid>https://speakerdeck.com/kannanmp/python-foundations-lect-1b-competitive-cybersecurity-101</guid>
    </item>
    <item>
      <title>Introduction to CTF [Lect. 1a: Competitive Cybersecurity 101]</title>
      <description>This presentation serves as the foundational module for "Competitive Cybersecurity 101," a specialized curriculum designed to equip students at an international school in Singapore, with the skills necessary to compete in the National Cybersecurity Olympiad (NCO) and the International Cybersecurity Olympiad (ICO).

As the cybersecurity landscape rapidly evolves—particularly with the integration of AI into both offensive and defensive operations—grassroots, hands-on education is more critical than ever. This deck outlines the strategic pathway from beginner to competitor, defining the core mechanics of Capture The Flag (CTF) competitions. It introduces participants to beginner-friendly, gamified learning environments like picoCTF, while establishing a strict, non-negotiable framework for ethical hacking and responsible disclosure in accordance with the Computer Misuse Act.</description>
      <media:content url="https://files.speakerdeck.com/presentations/ecd1446e801f4521b155f962e50c5ef8/preview_slide_0.jpg?40093954" type="image/jpeg" medium="image"/>
      <content:encoded>This presentation serves as the foundational module for "Competitive Cybersecurity 101," a specialized curriculum designed to equip students at an international school in Singapore, with the skills necessary to compete in the National Cybersecurity Olympiad (NCO) and the International Cybersecurity Olympiad (ICO).

As the cybersecurity landscape rapidly evolves—particularly with the integration of AI into both offensive and defensive operations—grassroots, hands-on education is more critical than ever. This deck outlines the strategic pathway from beginner to competitor, defining the core mechanics of Capture The Flag (CTF) competitions. It introduces participants to beginner-friendly, gamified learning environments like picoCTF, while establishing a strict, non-negotiable framework for ethical hacking and responsible disclosure in accordance with the Computer Misuse Act.</content:encoded>
      <pubDate>Fri, 24 Jul 2026 00:00:00 -0400</pubDate>
      <link>https://speakerdeck.com/kannanmp/introduction-to-ctf-lect-1a-competitive-cybersecurity-101</link>
      <guid>https://speakerdeck.com/kannanmp/introduction-to-ctf-lect-1a-competitive-cybersecurity-101</guid>
    </item>
    <item>
      <title>Judgement Day: Benchmarking “Black Box” LLMs With Open Legal Datasets</title>
      <description>Presented at the Linux Foundation Open Source Summit India (June 16, 2026) by Kannan Murugapandian.

This talk introduces Statutory Myopia — a newly documented failure mode in Retrieval-Augmented Generation (RAG) systems, where LLMs over-rely on retrieved text and suppress their own pre-trained knowledge of conflicting or superseding legal rules.

Using an open evaluation harness built on the LegalBench dataset (3 doctrinal task subsets, 239 unique cases, 4,302 reasoning traces), three frontier models — GPT-5, DeepSeek R1, and Gemini 2.5 Pro — were benchmarked across No-RAG and Statute-Only RAG conditions. The results were counterintuitive: incomplete retrieval caused a 9 percentage point drop in accuracy globally (0.87 → 0.78), with a worst-case 21 pp decline in the Hearsay task when the superseding Crawford v. Washington ruling was withheld from the context.

The talk also covers the custom async evaluation harness with SHA-256 cryptographic caching for reproducibility, the blacklist protocol for controlled retrieval experiments, and implications for any hierarchical domain — including medical guidelines, API deprecation, and compliance specifications.</description>
      <media:content url="https://files.speakerdeck.com/presentations/885659c4ffdc4dfcaf66c71d2371bda8/preview_slide_0.jpg?39737205" type="image/jpeg" medium="image"/>
      <content:encoded>Presented at the Linux Foundation Open Source Summit India (June 16, 2026) by Kannan Murugapandian.

This talk introduces Statutory Myopia — a newly documented failure mode in Retrieval-Augmented Generation (RAG) systems, where LLMs over-rely on retrieved text and suppress their own pre-trained knowledge of conflicting or superseding legal rules.

Using an open evaluation harness built on the LegalBench dataset (3 doctrinal task subsets, 239 unique cases, 4,302 reasoning traces), three frontier models — GPT-5, DeepSeek R1, and Gemini 2.5 Pro — were benchmarked across No-RAG and Statute-Only RAG conditions. The results were counterintuitive: incomplete retrieval caused a 9 percentage point drop in accuracy globally (0.87 → 0.78), with a worst-case 21 pp decline in the Hearsay task when the superseding Crawford v. Washington ruling was withheld from the context.

The talk also covers the custom async evaluation harness with SHA-256 cryptographic caching for reproducibility, the blacklist protocol for controlled retrieval experiments, and implications for any hierarchical domain — including medical guidelines, API deprecation, and compliance specifications.</content:encoded>
      <pubDate>Wed, 17 Jun 2026 00:00:00 -0400</pubDate>
      <link>https://speakerdeck.com/kannanmp/judgement-day-benchmarking-black-box-llms-with-open-legal-datasets</link>
      <guid>https://speakerdeck.com/kannanmp/judgement-day-benchmarking-black-box-llms-with-open-legal-datasets</guid>
    </item>
    <item>
      <title>Cracking the Code: From Caesar Ciphers to WhatsApp Encryption</title>
      <description>Introductory presentation on Cryptography presented at the STEAM Club of an international school</description>
      <media:content url="https://files.speakerdeck.com/presentations/6aea7df1665e40fda18c671e2d85060b/preview_slide_0.jpg?39552656" type="image/jpeg" medium="image"/>
      <content:encoded>Introductory presentation on Cryptography presented at the STEAM Club of an international school</content:encoded>
      <pubDate>Fri, 29 May 2026 00:00:00 -0400</pubDate>
      <link>https://speakerdeck.com/kannanmp/cracking-the-code-from-caesar-ciphers-to-whatsapp-encryption</link>
      <guid>https://speakerdeck.com/kannanmp/cracking-the-code-from-caesar-ciphers-to-whatsapp-encryption</guid>
    </item>
    <item>
      <title>Energy Nexus: Next-Generation Solar Integration for Urban and Rural Reliability (NYAS Junior Academy)</title>
      <description>As artificial intelligence and new technologies emerge, there is a rapidly increasing global demand for electricity. However, many regions cannot handle large amounts of renewable energy due to the surface area limitations and the 33% theoretical efficiency cap of current solar models. This presentation outlines our proposed solution to these challenges, developed as part of the New York Academy of Sciences (NYAS) Junior Academy Innovation Challenge.

Our research hypothesis explores the integration of Building-Integrated Photovoltaics (BIPV) for urban environments and multi-layer floating solar systems for water bodies. By incorporating advanced materials—such as Luminescent Solar Concentrator (LSC) technology and wide-bandgap semiconductors—alongside thermochemical storage, our hybrid solar systems are designed to achieve significantly higher energy reliability and efficiency than conventional solar frameworks.

Key Highlights of Our Research:

1. AI-Driven Performance Modeling: We fine-tuned an AI model using weather, irradiance, and environmental datasets to predict prototype performance across diverse real-world climates without needing a physical prototype.

2. Technology Readiness Validation: Our solutions are backed by existing literature, with validated high Technology Readiness Levels (TRL) in LSC technology, floating solar operations (like Singapore's Tengeh reservoir), underwater wide-bandgap PVs, and Calcium Looping Storage.

3. IP and Patent Analysis: We utilized AI-assisted patent analysis to ensure our combinations of materials and technologies are feasible, unique, and do not overlap with existing intellectual property.

Team Acknowledgments:
This project was a highly collaborative effort by Team Energy Nexus (Group #6310). I am incredibly proud of the research, engineering analysis, and dedication contributed by every member of our international team:

- Kannan Murugapandian (Singapore) [Team Lead]
- Liz Rodrigues (New Jersey, USA)
- Anadita Singh (Delhi, India)
- Agnibha Sengupta (West Bengal, India)
- Kashvi Ruparelia (Gujarat, India)
- Sreeja Padmala (Michigan, USA)

To view our full project portfolio and team background, please visit our team profile at JoinLaunchPad: https://joinlaunchpad.com/#/projects/6310/solar-energy-team-1

Note: Photos of my team members in the presentation slides have been replaced with standardized icons to respect their privacy and adhere to digital ethics, as this deck has been uploaded to a public platform.</description>
      <media:content url="https://files.speakerdeck.com/presentations/2bbb9c65465442b3a87b99b503c2405a/preview_slide_0.jpg?39079482" type="image/jpeg" medium="image"/>
      <content:encoded>As artificial intelligence and new technologies emerge, there is a rapidly increasing global demand for electricity. However, many regions cannot handle large amounts of renewable energy due to the surface area limitations and the 33% theoretical efficiency cap of current solar models. This presentation outlines our proposed solution to these challenges, developed as part of the New York Academy of Sciences (NYAS) Junior Academy Innovation Challenge.

Our research hypothesis explores the integration of Building-Integrated Photovoltaics (BIPV) for urban environments and multi-layer floating solar systems for water bodies. By incorporating advanced materials—such as Luminescent Solar Concentrator (LSC) technology and wide-bandgap semiconductors—alongside thermochemical storage, our hybrid solar systems are designed to achieve significantly higher energy reliability and efficiency than conventional solar frameworks.

Key Highlights of Our Research:

1. AI-Driven Performance Modeling: We fine-tuned an AI model using weather, irradiance, and environmental datasets to predict prototype performance across diverse real-world climates without needing a physical prototype.

2. Technology Readiness Validation: Our solutions are backed by existing literature, with validated high Technology Readiness Levels (TRL) in LSC technology, floating solar operations (like Singapore's Tengeh reservoir), underwater wide-bandgap PVs, and Calcium Looping Storage.

3. IP and Patent Analysis: We utilized AI-assisted patent analysis to ensure our combinations of materials and technologies are feasible, unique, and do not overlap with existing intellectual property.

Team Acknowledgments:
This project was a highly collaborative effort by Team Energy Nexus (Group #6310). I am incredibly proud of the research, engineering analysis, and dedication contributed by every member of our international team:

- Kannan Murugapandian (Singapore) [Team Lead]
- Liz Rodrigues (New Jersey, USA)
- Anadita Singh (Delhi, India)
- Agnibha Sengupta (West Bengal, India)
- Kashvi Ruparelia (Gujarat, India)
- Sreeja Padmala (Michigan, USA)

To view our full project portfolio and team background, please visit our team profile at JoinLaunchPad: https://joinlaunchpad.com/#/projects/6310/solar-energy-team-1

Note: Photos of my team members in the presentation slides have been replaced with standardized icons to respect their privacy and adhere to digital ethics, as this deck has been uploaded to a public platform.</content:encoded>
      <pubDate>Tue, 14 Apr 2026 00:00:00 -0400</pubDate>
      <link>https://speakerdeck.com/kannanmp/energy-nexus-next-generation-solar-integration-for-urban-and-rural-reliability-nyas-junior-academy</link>
      <guid>https://speakerdeck.com/kannanmp/energy-nexus-next-generation-solar-integration-for-urban-and-rural-reliability-nyas-junior-academy</guid>
    </item>
    <item>
      <title>Introduction to Digital Forensics</title>
      <description>"Every contact leaves a trace." In this presentation designed for junior students at DPS International School, I break down the fundamentals of Digital Forensics within the context of Capture The Flag (CTF) cybersecurity competitions.

The goal of this session was to introduce students to the methodology of digital crime scene investigation, equipping them with the practical tools and ethical guidelines needed to solve forensic challenges.</description>
      <media:content url="https://files.speakerdeck.com/presentations/1c8699fee7e24a2f8a1f25f7bf18ae8e/preview_slide_0.jpg?38908481" type="image/jpeg" medium="image"/>
      <content:encoded>"Every contact leaves a trace." In this presentation designed for junior students at DPS International School, I break down the fundamentals of Digital Forensics within the context of Capture The Flag (CTF) cybersecurity competitions.

The goal of this session was to introduce students to the methodology of digital crime scene investigation, equipping them with the practical tools and ethical guidelines needed to solve forensic challenges.</content:encoded>
      <pubDate>Sun, 29 Mar 2026 00:00:00 -0400</pubDate>
      <link>https://speakerdeck.com/kannanmp/introduction-to-digital-forensics</link>
      <guid>https://speakerdeck.com/kannanmp/introduction-to-digital-forensics</guid>
    </item>
    <item>
      <title>Democratizing Personalized Learning with Open Source AI</title>
      <description>How do we solve Bloom’s 2 Sigma Problem and scale highly effective 1-to-1 tutoring to every student? While Artificial Intelligence offers a proven solution—significantly boosting student performance and reducing teacher planning hours by 40%—relying on closed-source AI models introduces massive risks.

In this presentation from the FOSSASIA 2026 Summit, I explore why the future of personalized education must be open-source, and how we can architect AI systems that teach students how to think, rather than just giving them the answers.

Talk: https://youtu.be/1CHsfYtQ_kw</description>
      <media:content url="https://files.speakerdeck.com/presentations/bf8409ceea70434095c95382c43438cb/preview_slide_0.jpg?38908462" type="image/jpeg" medium="image"/>
      <content:encoded>How do we solve Bloom’s 2 Sigma Problem and scale highly effective 1-to-1 tutoring to every student? While Artificial Intelligence offers a proven solution—significantly boosting student performance and reducing teacher planning hours by 40%—relying on closed-source AI models introduces massive risks.

In this presentation from the FOSSASIA 2026 Summit, I explore why the future of personalized education must be open-source, and how we can architect AI systems that teach students how to think, rather than just giving them the answers.

Talk: https://youtu.be/1CHsfYtQ_kw</content:encoded>
      <pubDate>Sun, 29 Mar 2026 00:00:00 -0400</pubDate>
      <link>https://speakerdeck.com/kannanmp/democratizing-personalized-learning-with-open-source-ai</link>
      <guid>https://speakerdeck.com/kannanmp/democratizing-personalized-learning-with-open-source-ai</guid>
    </item>
  </channel>
</rss>
