LLama2.java: LLM integration with A 100% Pure Java file

An airhacks.fm (https://airhacks.fm) conversation with Alfonso Peterssen (@TheMukel (  / themukel  ) ) about:
discussion about Alfonso's early programming experience and participation in the IOI (https://stats.ioinformatics.org) competition, studying computer science and functional programming with Martin Odersky, internships at Google and Oracle Labs (  / oracle_labs  ) working on compilers and the Espresso (https://www.graalvm.org/latest/refere...) project implementing a JVM in Java,
espresso mentioned in "#208 GraalVM: Meta Circularity on Different Levels" (https://airhacks.fm/#episode_208) , "#194 GraalVM, Apple Silicon (M1) and Clouds" (https://airhacks.fm/#episode_194) , "#167 GraalVM and Java 17, Truffle, Espresso and Native Image" (https://airhacks.fm/#episode_167) and "#157 The Ingredients of GraalVM" (https://airhacks.fm/#episode_157) ,
porting LLVM (https://llvm.org) to pure Java in one class, integrating Large Language Models (LLMs) in Java by porting the LLAMA model from C to Java,
GPU acceleration with tornadovm (https://www.tornadovm.org) ,
TornadoVM appeared at "#282 TornadoVM, Paravox.ai: Java, AI, LLMs and Hardware Acceleration" (https://airhacks.fm/#episode_282) ,
performance of the Java port being within 10% of the C versions, potential huge opportunities for integrating AI and LLMs with enterprise Java systems for use cases like fraud detection, the Java port being a 1,000 line self-contained implementation with no external dependencies, the need for more resources and support to further develop the Java LLM integration,
the llama2.java (https://github.com/mukel/llama2.java) project


Alfonso Peterssen on twitter: @TheMukel (  / themukel  )