While in Pt.1, we directly presented Cognify’s optimization result of the 4-step workflow, in this blog post, we dive deeper into the technical secret sause behind Cognify’s efficient and effective autotuning: the AdaSeek algorithm, a novel adaptive hierarchical Bayesian-Optimization search algorithm. AdaSeek is result driven, efficient, budget-aware, and has wide search space coverage.

Is it possible to autotune a 4-step gen-AI workflow's generation quality with a budget of $5 and 30 minutes instead of $168K and weeks as in a naive tuning approach? In this blog post, we will walk through how Cognify autotunes such a workflow with three types of optimization techniques: those that change the architecture of a workflow, those that improve single steps in a workflow like model selection, and those that tweak individual steps' input like prompts.

Building high-quality, cost-effective generative AI applications is challenging due to the absence of systematic methods for tuning, testing, and optimization. We introduce Cognify, a tool that automatically enhances generation quality and reduces costs for generative-AI workflows, including those written with LangChain, DSPy, and annotated Python. Built on a novel foundation of hierarchical, workflow-level optimization, Cognify delivers up to a 48% improvement in generation quality and up to 9x cost reduction. Cognify is publicly available at https://github.com/WukLab/Cognify.