-|{: style="margin:0em; padding:0em; width:10em"}|
+|{: style="margin:0em; padding:0em; width:10em"}|
| **Steffen Illium** *AI Consultant & Data Scientist*|
-[Grab my CV here](\assets\illium_cv_censored.pdf){: .btn .btn--success}
+[Grab my CV here](/assets/illium_cv_censored.pdf){: .btn .btn--success}
-My academic and professional path reflects a deep-seated interest in transforming data into actionable insights, beginning with a foundation in Geography (BSc, JGU Mainz) and Geo-Informatics (MSc, University of Augsburg), and culminating in a PhD in Computer Science from LMU Munich (summa cum laude). During my doctoral studies and subsequent research assistant role at LMU (2018-2023), I focused on advancing machine learning models for sequential data, self-learning systems, and contributing to foundational research in neural network applications.
+My career has been driven by a passion for turning data into actionable insights. This journey began with a BSc in Geography from JGU Mainz and an MSc in Geo-Informatics from the University of Augsburg, culminating in a PhD in Computer Science (summa cum laude) from LMU Munich. During my doctoral studies and subsequent role as a research assistant at LMU (2018-2023), I focused on developing machine learning models for sequential data, advancing self-learning systems, and contributing to foundational research in neural network applications.
-My research frequently involved collaborations with industry partners on projects such as "ErLoWa" (leak detection in water networks) and "AI-Fusion" (emergent dysfunction detection in MA-RL), providing extensive experience in areas like audio signal processing, deep learning for sequence and image analysis, and multi-agent reinforcement learning (MARL), particularly concerning safety and emergence in industrial contexts. This blend of theoretical research and practical application forms the basis of my [publications](/publications) and [research](/research) activities.
+A significant part of my research involved collaborating with industry partners. Key projects include "ErLoWa," focused on leak detection in water networks, and "AI-Fusion," which dealt with detecting emergent dysfunctions in multi-agent reinforcement learning (MA-RL). These experiences provided me with extensive expertise in audio signal processing, deep learning for sequence and image analysis, and MARL, particularly concerning safety and emergence in industrial contexts. This blend of theoretical research and practical application is detailed in my [publications](/publications) and [research](/research) pages.
-Beyond core research, I have actively engaged in teaching and academic service. My experience includes lecturing, supervising practical courses (e.g., iOS, Android development), managing seminars (IMAPS), leading Python crash courses, and mentoring numerous Bachelor's (20) and Master's (9) theses. Details can be found on the [teaching](/teaching) page.
+Beyond my core research, I have actively contributed to teaching and academic service. My experience includes lecturing, supervising practical courses (such as iOS and Android development), managing seminars (IMAPS), leading Python crash courses, and mentoring numerous students through their Bachelor's (20) and Master's (9) theses. You can find more details on my [teaching](/teaching) page.
-Additionally, I have embraced leadership and organizational roles within the academic community. I served as the lead organizer for the [OpenMunich conference](https://openmunich.eu) (2018-2019) and headed the editorial team for the ["DIGITALE WELT Magazin"](https://digitaleweltmagazin.de/) (2018-2023), broadening my experience in project management, communication, and community building.
+I have also embraced leadership and organizational responsibilities within the academic community. I served as the lead organizer for the [OpenMunich conference](https://openmunich.eu) from 2018 to 2019 and headed the editorial team for ["DIGITALE WELT Magazin"](https://digitaleweltmagazin.de/) from 2018 to 2023. These roles broadened my experience in project management, communication, and community building.
---
@@ -90,8 +90,6 @@ Additionally, I have embraced leadership and organizational roles within the aca
-
-
**Libraries & Frameworks (Python Focus):**
@@ -102,6 +100,10 @@ Additionally, I have embraced leadership and organizational roles within the aca
+
+
+
+
**Systems & DevOps:**
diff --git a/assets/illium_cv_censored.pdf b/assets/illium_cv_censored.pdf
index 022b0635..cd51911f 100644
Binary files a/assets/illium_cv_censored.pdf and b/assets/illium_cv_censored.pdf differ
diff --git a/nginx_default.conf b/nginx_default.conf
index e4fb7a10..abda72c9 100644
--- a/nginx_default.conf
+++ b/nginx_default.conf
@@ -7,7 +7,8 @@ map $http_accept $webp_suffix {
# Map to capture the image path *without* the file extension
# This regex captures everything before the last dot and jpg/jpeg/png extension
map $uri $image_path_without_ext {
- ~^(?.+)\.(?:jpe?g|png)$ $captured_path;
+ default ""; # Good practice to have a default
+ ~*^(?.+)\.(?:jpe?g|png)$ $captured_path; # Use ~* for case-insensitive extension match
}
server {
@@ -29,11 +30,6 @@ server {
add_header Cache-Control "public";
# Try to serve the .webp file first if browser supports it
- # $image_path_without_ext comes from the map above
- # $webp_suffix comes from the map above (.webp or empty)
- # If $image_path_without_ext$webp_suffix exists (e.g., /path/image.webp), serve it.
- # Otherwise, try the original $uri (e.g., /path/image.png).
- # If neither exists, return 404.
try_files $image_path_without_ext$webp_suffix $uri =404;
}
@@ -54,7 +50,7 @@ server {
internal;
}
- # Gzip settings (keep as they are)
+ # Gzip settings
gzip on;
gzip_comp_level 4;
gzip_types text/html text/plain text/css application/json application/x-javascript text/xml application/xml application/xml+rss text/javascript;