Email the Author
You can use this page to email Isaak Tsalicoglou about Elixir Chatbot Alchemy.
About the Book
Following the same engineering ethos and exploratory learning approach that proved so popular with Northwind Elixir Traders, Elixir Chatbot Alchemy explores the implementation of a chatbot in Elixir. By utilizing the REST API of a self-hosted instance of Ollama and CPU inference on small(-ish) open-weight LLMs, we explore how LLMs operate, the impact of model choice on the quality of inference, and use Phoenix LiveView to build a variety of features, such as document summarization and a chatbot that can be used on a webpage, and an admin UI with reporting of the use of the chatbot.
Note: this is a super-early low-fanfare release, subject to heavy changes of scope and content.
About the Author
Spurred by a serendipitous discussion with a good friend on the merits of functional programming, Isaak has since 2022 focused his programming-related activities exclusively on learning and using Elixir to develop further great software for others, as well as for his own use as an "indiehacker" by combining full-stack development skills with his extensive product development, management and marketing experience.
Isaak is an ETH Zürich Mechanical Engineer with an MBA from IMD, Switzerland. His software, training and consulting work builds on three decades of experience in using computers and programming to turbocharge products, services, and operating business models. An avid fan of open-source software and self-hosting, an early adopter of Linux since 1997, and a relentless tinkerer with NetBSD and FreeBSD, Isaak has worked with Fortran (77, 90), MATLAB, GNU Octave and Mathematica for scientific simulations, C for mobile robotics, basic C++ for refactoring proprietary FEA post-processing code, and a lot of Python for engineering optimization, neural networks and ensembles, machine learning, Monte Carlo simulations, CAD automation, and FEA and CFD model generation. He has also worked with PHP for web-based software and websites, Go for simple tasks, as well as built REST APIs for machine learning predictions of engineered components and the natural-language processing of web-scraped data, and REST APIs with simple web UIs for managing internal and customer-facing processes of an industrial equipment trading business.