My new chatbot is online! I opted for the Software-as-a-Service solution from Botsonic for the implementation. You can read all the details about the implementation and why I chose Botsonic in this article.

---

To repeat

What are LLM chatbots?

LLM chatbots are computer programs that use machine learning and natural language processing to simulate human interactions in order to understand language, generate natural responses and have human-like conversations. LLM chatbots use linguistic computing to create a large language model (LLM) that can generate anything from text to images to music based on user prompts. You can read more about LLM chatbots in my article LLM chatbots – An introduction to the new world of bots.

What are the advantages of LLM chatbots?

LLM chatbots offer companies and end users a wide range of benefits. Below are some general advantages of chatbots, followed by specific advantages of LLM chatbots compared to chatbots without LLMs:

1. Availability
2. Scalability
3. Cost efficiency
4. Consistency
5. Data collection

Advantages of LLM chatbots over traditional chatbots:

1. Improved understanding
2. Adaptability
3. Personalization
4. Ability to generate long texts
5. Integration of external knowledge

You can find more explanations about the advantages of LLM chatbots in my article LLM chatbots – An introduction to the new world of bots.

---

Why do I need a new LLM chatbot?

As a consultant for generative AI as well as chatbots and voicebots, I have supervised a large number of customer projects in the field of LLM chatbots and gained extensive experience in the process. This experience was both technical and conceptual in nature and primarily concerned the implementation phase. I have also repeatedly observed how companies have benefited from the new LLM chatbots. Customers began to use the LLM chatbots with enthusiasm and often achieved impressive results. It was particularly remarkable when the LLM chatbots not only answered simple customer queries, but also clarified complex issues to their satisfaction or even took on up-selling tasks.

These findings and the increasing generation of customer inquiries via my own website prompted me to develop my own LLM chatbot.

I pursue the following main goals with my own LLM chatbot:

1. Automated answers to questions about chatbots, artificial intelligence and my services, around the clock.
2. Ensuring that no promising customer inquiries are overlooked by having the bot forward important inquiries directly to me.
3. Expanding my experience in the field of LLM chatbots and getting to know new application possibilities.

How do I find the right technology for my LLM chatbot?

When looking for the right LLM chatbot software for my own website, I proceeded as described in my article Finding LLM chatbot software.

I first defined my goals, needs and expectations and then started looking for a suitable provider. I focused on the following criteria:

1. Simplicity of the solution in terms of customization of system prompts, design and data sources
2. Costs
3. Contract conditions and termination modalities
4. Accessibility of customer support
5. Integration options in other communication channels, such as Instagram and WhatsApp

Why did I implement my LLM chatbot with Botsonic?

Thanks to my various LLM chatbot projects with customers, I already know a large number of LLM chatbot software providers. I quickly realized that the SaaS solution from Botsonic seemed to be the most suitable for me and my website at the time.

1. very simple operation

The Botsonic solution is extremely easy to use. The following steps are necessary to implement the first version of the new LLM chatbot:

• Enter a website URL as a data source or upload other knowledge documents
• Addition of system prompts in ready-made prompt templates
• Customization of the chatbot greeting and other help texts
• Definition of the chatbot design or avatar
• Selection of available languages
• Testing
• Inserting the HTML code on your own website or linking to other communication channels

2. good price-performance ratio

Botsonic initially offers a complete free version for the first two weeks for testing purposes. After that there are the following price packages.

I am currently using the cheapest price package and am very satisfied with the range of functions and the price-performance ratio.

Tip: Send me a short message and I can send you a discount code for Botsonic.

3. great customer service

As part of my connection to Instagram and GoogleMyBusiness, I initially had a few questions and challenges. So I contacted Botsonic’s customer service and had an answer within six hours, including a reply. Video tutorial. The customer service advisor responded very individually to my questions and was able to overcome all challenges in a short time.

4. many integration options with other communication channels

Anyone who knows me knows that I’m a big fan of WhatsApp and have recently started using Instagram more and more. That’s why it’s also very important to me to be able to integrate automation into these communication channels in the future. Botsonic offers a wide range of integration options in other channels and also has many other interfaces. In the end, this criterion was very decisive for me when I opted for Botsonic instead of other cost-effective SaaS solutions.

What were the challenges in implementing my LLM chatbot?

The implementation with Botsonic was extremely straightforward and presented me with hardly any unknown challenges. Of far greater importance, however, was the development of my own LLM chatbot concept and the precise formulation of the system prompts.

First of all, I had to carefully weigh up how much responsibility I wanted to give my new LLM chatbot and which dialogs and customer inquiries I would continue to answer myself in future. These considerations then had to be formulated in suitable prompts. I also created additional prompts to define the behavior of my LLM chatbot in different situations.

Conclusion: My LLM chatbot with Botsonic

After about three days (including breaks), the first version of my LLM chatbot was finished. Adding the knowledge base by entering the URL of my website was almost automatic. The design adjustments and settings for welcome and information texts also took comparatively little time. Creating your own prompts, including a well thought-out concept, was more time-consuming.

This was followed by a short test phase, carried out by myself as well as friends and family members. Based on the test results, I made a few adjustments to the prompts and reduced the size of the knowledge base. The LLM chatbot no longer uses the entire content of my website, but only selected pages.

The LLM chatbot can now be tested via the chat window on my website.

And when do you start?

Would you like to start your own LLM chatbot project? Or do you have any other questions first?

In both cases you are very welcome to contact me. I have already accompanied and managed many LLM-Chatbot projects and am very happy to be part of your project as well.

Just send me a message – preferably via WhatsApp message or e-mail.

This article is also available as a podcast episode

Attention! The podcast was created entirely by my AI-Assistant based on my contribution – no guarantee for incorrect content.

*I used SwissGPT’s AI technology to optimize the language of this article.