Tag
DSPy 3.4.0 has been released, featuring native support for Jev and System One models along with a new optimizer, ReAnchor, for calibrating outputs with confidence.
DSPy 3.4.0 is released with native support for Jev and System one models and a new optimizer, ReAnchor, for calibrating outputs with confidence.
This article describes using DSPy code to generate choice-based JEV decision training data from raw text, highlighting the cost-effectiveness with cheap models like gpt-6-luna and deepseek-v4.1-flash.
ClusterFewshot is a novel method for improving few-shot demonstration selection in LLM workflows by integrating semantic clustering and utility scoring, which reduces optimization costs and enhances accuracy in DSPy-based pipelines.
The tweet emphasizes the importance of careful problem specification in machine learning over attachment to methods, echoing DSPy's philosophy and referencing Sutton's bitter lesson.
This podcast episode discusses AI tokenomics and the emerging issue of 'tokenflation', focusing on measuring the value of AI tokens, the limitations of benchmarks, and future innovations in AI efficiency.
A patch release for DSPy includes CodeInterpreter fixes, MCP 2.0 support, GEPA bump, and improvements to errors and callbacks.
Introduces Flex, a new DSPy module that lets language models rewrite the program code itself rather than just prompts, enabling better optimization, fewer model calls, and safer execution via sandboxing.
Flex is a new DSPy feature where optimizers can replace LLM calls with deterministic code, sometimes improving accuracy while drastically reducing the number of LLM calls needed.
DSPy 3.3.0 is released, featuring dspy.Flex for GEPA-optimized code and prompts, and dspy.ReActV2 with native parallel tool calling, with praise for lead maintainer Isaac Miller.
Using the DSPy framework, the author evaluates and improves the system prompts for Datasette Agent's SQL query feature, identifying issues such as column-name guessing and error-retry loops.
Maxime Rivest announces a main stage keynote for DSPy at the AI Engineer conference, aiming to grow the tool's accessibility and invites feedback.
A tweet recommends watching Angie Jones' keynote on building AI agents, highlighting core engineering principles and mentioning Python frameworks like DSPy, Functai, and Mirascope.
A practical guide explaining three levels of building self-improving AI agents, from manual loops to automated design, with recommended tools and frameworks.
Discussing a figure from an essay on AI's impact on software engineering that highlights irreducible complexity and predicts evolution of software engineering, with DSPy cited as an early example.
The article explains how GEPA (Genetic-Pareto Optimization) within DSPy is used for efficient prompt tuning, specifically applied to pretraining data curation at Microsoft AI, allowing researchers to replace manual prompt engineering with automated compute-driven optimization.
GEPA-optimized LLM judges from dspy are used for data filtering in Microsoft's MAI-Thinking-1 model pre-training pipeline.
The author compares Structural Equation Modeling, Neural ODEs, and AI Programs like DSPy as declarative frameworks for defining and optimizing computational graphs, arguing that structured flows are essential for trustworthy AI agents.
Ax is an open-source TypeScript library that implements DSPy-style typed signatures and agent frameworks for building reliable AI applications with minimal prompting. It supports multiple LLM providers and includes features like agents, flows, RAG, and self-improving pipelines.
DSPy has a new front page and documentation for easier onboarding, and is approaching a major DSPy 4.0 release with radical new ideas.