Skip to content

Category: Ohne Kategorie

Monatsupdate August

Dissertation Update

Updates an der Publication-Front:

Das Paper Integration of Metaverse and Multimedia Information Retrieval von der ICSA ist nun in der ACM Digital Library verfügbar

Zum 256 Metaverse Recording Dataset habe ich ein Paper geschrieben, dass auf der ACM Multimedia 2024 angenommen wurde! Hier gehts zum OpenReview

Mit einem Studenten habe ich ein Paper zur Avatar Erkennung geschrieben. Dies ist als Preprint verfügbar.

Research Topics:

Image generation: Aktuell exploriere ich mit Bildgenerierung von Metaverse Daten für die Result Presentation. Experimente laufern mit Stable Diffusion 3 (medium und large). Beide Modelle sind sehr interessant, vor allem wiel die Textgenerierung in den Bildern schon sehr gut läuft, wenn auch nicht perfekt.

Weitere Experimente laufen mit Obejct Detection und Sprachanalyse. Detectron 2 bietet eine gute Umsetzung von R-CNN Varianten, eine Alternative zum bekannteren YOLO Netzen. Auch die neuen YOLO Versionen (v10 , April oder Mai veröffentlicht) wäre eine Untersuchung auf Metaverse Content Wert. Also auf die Todo Liste. Dazu gehe ich in die Sprachanalyse, ich bin sher begeistert von der Qualität von Whisper und untersuche nun die Toxizität in den Metaverse Aufnahmen.

 

Skill2Lead

Skill 2 Lead: Ich arbeite weiter an Coaching Inhalten für Fachkräfte, die sich zur Führungskraft entwickeln wollen oder es jüngst wurden. Als Sideproject aktuell nicht die Top-Prio.

Sonstiges

Juli: 145,6 km Rad gefahren. Nicht sooo viel, aber immherin summiert es sich auf 638,2 km in diesem Jahr. Wird mal wieder Zeit für 3-stellige Touren :)

VeloWear App: Kann man mit KI eine App bauen, ohne Programmierkenntnisse zu haben? Ja, soweit geht das schon. Ich habe den ersten Teil einer YT-Serie gestartet. Die App ist schon im Testflight Mode, Part 2 kommt bald.

1 Comment

Updates KW 25

Hey Leute, vergangene Woche war vollgepackt mit spannenden Projekten und Aktivitäten. Hier ein kurzer Überblick:

Tech-Projekte:

  1. RAG mit Langchain: Ich habe endlich Zeit gefunden, Retrieval-Augmented Generation (RAG) mit Langchain zu testen. Die Evaluierungen waren recht aufschlussreich, und ich bin begeistert von den Möglichkeiten, die sich dadurch für meine Projekte eröffnen.
  2. llama.cpp: Parallel dazu habe ich mich mit llama.cpp beschäftigt. Es ist faszinierend zu sehen, wie effizient diese Implementierung große Sprachmodelle auf Consumer-Hardware laufen lässt.
  3. Whisper OpenAI Plugin: Ein echtes Highlight war die Integration des Whisper OpenAI Plugins für Information Retrieval. Ich habe damit gleich ein 256 Metaverse Dataset indexiert – die Ergebnisse sind vielversprechend!

Sonstiges:

  • Sport: Trotz des vollen Terminkalenders habe ich es geschafft, regelmäßig Sport zu treiben. Es hilft ungemein, den Kopf frei zu bekommen.
  • Kleinanzeigen: Ich habe mich endlich dazu durchgerungen, ein paar Sachen zu verkaufen, die ich nicht mehr brauche. Überraschend, wie viel sich da über die Zeit ansammelt!

 

Das war’s erstmal von mir. Wie war eure Woche so? Bis bald!

Leave a Comment

Bucket List Achievement: My first scientific paper

I am beyond excited to share with you all a significant milestone in my academic journey: the publication of my first scientific paper in the MDPI Big Data and Cognitive Computing Magazine! As someone deeply passionate about the fields of computer science, this accomplishment is not only a testament to my dedication, but also a realization of a long-time goal. In this blog post, I’d like to share the story behind my research and the path that led to this achievement.

Leave a Comment

My 2022 in review

The end is near, I mean the end of the year. I want to look back and reflect on my personal year.

Achievements

First public conference talk(s): At this year’s FrOSCon, I was invited to give two conference talks – on one day. That was my first time at a public conference and it went very well.

Published first print article: I could land a spot in a German IT magazine, “IT-Spektrum”. It was a long dream of mine to do this and I was very happy that it worked on the first try.

Published a web article on GraphQL: Together with Oliver I wrote an article about GraphQL Federation in microservice architectures.

Leave a Comment

4 Tips for development of Alexa Skills

In the last weeks, I have developed some Alexa Skills for different purposes. It is really cool to develop the skills with the Alexa developer console. Building and testing the dialogue model is fairly easy. But at some points, you may encounter some problems like me. Therefore, I would like to share some tips with you to improve the user experience of a skill significantly.

Tip 1: Use Default Slot Types

Let’s start with a simple topic. If possible, use the provided slot types from Amazon, like Amazon.FOOD or AMAZON.NUMBER. These Slots have a huge set of data in the background. They are already optimized for a good NLP understanding. Doing this on your own is a lot of work and fine-tuning the model. Save yourself many hours and use what Amazon provides you.

Tip 2: Use a proxy for local development

There are different ways to implement the logic for the service: AWS Lambda or (self-hosted) endpoint services. If you develop endpoints services, you need to redirect the requests from the Alexa skill to the development instance, usually running on the local machine. An important thing is, the service needs to provide a valid TLS certificate. The easiest way to get it running is a web-proxy system like ngrok. Ngrok routes requests via a public web URL to your local development instance. And the best thing is, it has an option to provide a valid Wildcard-TLS-Endpoint which will be accepted by Alexa. This saves you a heck of time to set up anything equivalent with DynDNS and creating certificates. ngrok - a good tool for developing Alexa Skills

Tip 3: Answer not only use-case questions

During the development of Alexa skills, you work a lot through the questions (utterances) you have in mind with regard to the use case. But, think about your users. They can just interact with your app by asking questions. They can not click through a mobile app or website to search and find things they need. It’s important to be prepared for simple and general questions such as:

  • “What are the opening hours?”
  • “What is the address of a store?”
  • “What is the maximum of items I can order?”

Think about how your customer will ask questions. Ask your friends to try the skill and listen to their natural type of questions and commands. You can also log questions in the FallbackIntent to find out what real people say.

Tip 4: Test Alexa Skill dialogue with many people

This tip continues the thoughts of the previous. Many people will formulate questions and commands differently. Since the skill is usually used by many people, you need to be prepared for different types of utterances. Add as many sample utterances as you can to improve the user experience for the skill.

These 4 tips will improve the user experience of your Alexa skill. Do you have any further tips? Let me know in the comments.

2 Comments

Recap of one year at tarent solutions

Start of a journey

I started as Head of Emerging Technologies at tarent solutions in September 2018. After a year well spent, it was time to draw a conclusion. As part of Business Development, the Emerging Technologies division is responsible for identifying relevant technologies for our customers and expanding the company’s existing know-how with new technologies. As a company for individual software development projects, tarent’s strength lies in its mastery of current technologies in software development: early in agile, explorers of microservice architectures, first to use the GO programming language productively. The focus in the first year was on IoT & KI.

Leave a Comment

Retrospective 12/2017 & Year Review 2017

Retrospective 2017

Happy New Year 2018! I hope everybody had a vibrant start to the new year. Like the end of last year, I updated my goals and set up new goals for 2018 in the last days. As usual, the main problem is the lack of time. Here and there are some possible obstacles in sight, but to achieve all I want to do I need more time. You know? At least with my (new) job with the Edge Computing, I can enjoy technology & programming again. I hope you will stay with me the next year. I plan some more blog posts about the Edge stuff. But, I won’t promise anything, because, you know… procrastinating and so on. But let me recap:

What have I done in December

  • finalized projects for CoT – there are some nice updates in the pipe for early 2018
  • completed my overall 2017 cycling goal
  • started my first experiments on the DT edge computing platform – I’m
  • submitted a proposal for buildingIoT 2018
  • preparations for MWC 2018
  • Christmas shopping ordering

Thought

  • maybe I should take more pictures again
  • it needs a clear definition of Fog and Edge Computing
  • these books don’t write themselves
  • this could be a sustainable business…

Read

 

Recap of 2017

Let me start off by thanking everybody who was with me in 2017 and for all your support!

  • Business: in 2017, there were more up’s and down’s than usual. But since Cloud of Things was migrated to our superior hosting partner, we had an uptime of 100%. Since then we had a dry period. Some maintenance and organizational changes, but no surprises. The numbers were sophisticating. In the second half, there was a hard time. Setting up new initiatives but still, do the right things.
  • Private goals: there are some checked and a few missed goals. Checked? Definitely sports
veloviewer-infographic
Veloviewer Infographic of my activities in 2017

Let’s start strong in 2018 and thank you for being here!

1 Comment

IoT is dead, long live Edge Computing

In the last 3.5 years, I was responsible for the service development of the Cloud of Things, the IoT platform of Deutsche Telekom. It’s been an incredible adventure, but after almost four years, it’s time for me to move on. Developing a product for the Internet of Things was a fantastic adventure. Learning the various different customer problems and build a full-service stack to solve them with ease, fast and secure. We started from a small base and have grown up to a major pillar in the DT IoT strategy. But, inside Deutsche Telekom, the topic went “big”, too big. I selected my next adventure: low latency edge computing. The Low Latency Computing platform is built to reduce the latency between the computing device and the cloud. The Low Latency Computing platform enables customer solutions like suitable Car-to-X communication. It helps to smooth the end-customer experience with low latency, e.g.…

1 Comment

Retrospective 11/17

Retropective-11-17 - Chris Lawton

A lot happened in the past months. There was a big change in the company regarding IoT. A re-organization, new goals, a broader approach to the market. With little free time, I couldn’t manage to write about the details, I just tweeted about short stuff. But now, by the end of the year, I find time to write some updates.

What have I done:

  • managed, managed and managed – less real work
  • we released version 1.0 of the Cloud of Things developer SDK
  • I voted in the general election
  • changed my job – more on this in another post
  • starting development of some features on Cloud of Things and
  • start writing a book

Thought:

  • about platform strategy, technology development
  • much about my job and what I want to do
  • (1 try) change it, (2) leave it, love it
  • how to live more healthy and mindful
  • write a book!

Read:

1 Comment

A short recap of the missing half of 2016

This is a dirt bike!
This is a dirt bike!

Maybe you are a regular reader of my blog. By regular I mean at least once a year or so. ;-) If so, you may have noticed the last half of the year this blog was more or less abandoned. But this was not because nothing happened. It happened so much that I didn’t find the time to write. A common thing I’m sure you know about. But was going on? 

I took part in two Enduro Mountain Bike races. It was such a great experience. I got much impressions, learned to handle my bike in deep mud. Okay, I don’t got it, I crashed several times. I got a glimpse of what it means to be constant during a season. Daily training eats up your time, that’s for sure. Keeping up the motivation after three months was too hard for me. Results? I don’t want to talk about ;-), but I had great fun at the races and that’s what counts for me.

I have married! As usual, I keep the private things private on this blog. I decided to take over my wife’s surname. I updated this site domain, so please update your bookmarks, RSS readers, etc. to patricksteinert.de for best reading experience.

Leave a Comment