Preference Tuning LLMs: PPO, DPO, GRPO — A Simple GuideWith all the hype surrounding DeepSeek and the innovative methods their researchers have introduced, reinforcement learning applied to LLM…Feb 22Feb 22
Building a Sandboxed Environment for AI generated Code ExecutionWhen building AI agents, especially those that dynamically generate and execute analysis code, one major concern is security and stability…Jan 7Jan 7
CSV Agent for Enhanced Data InteractionGet plots and more with your csv agentApr 25, 2024Apr 25, 2024
Exploring DSPy — Structure in the Chaos of LLM Programming?Imagine you’re trying to have a more meaningful conversation with a smart robot using only a set of pre-written cue cards. You might find…Mar 3, 2024Mar 3, 2024
Langchain CSV Agent — a chain of function calls (Part 1)Autonomous agents powered by LLMs are said to be the future and a subject of much enthusiasm as well as spook at the moment. While they…Oct 29, 2023Oct 29, 2023
Published inGoPenAIEarly Reflections on the Potential and Limitations of LLMs in productionFOMO kicked in, and I finally got around to building a chatbot that lets you ‘chat’ with your custom documents. If you are a developer…Apr 14, 2023Apr 14, 2023
Published inTDS ArchiveSemantic Similarity NetworksMining knowledge from inter-connectedness of documentsSep 23, 20221Sep 23, 20221
Data quality for application developmentData quality is important for modern application/software development for a variety of reasons. First, data is the lifeblood of any…Jul 26, 2022Jul 26, 2022
