DSPy Logo

DSPyWeekly

Your Weekly Dose Of All Things DSPy

Articles and Tutorials
Videos
Every Friday

You'll receive verification email in max 1 minute. Do check SPAM just in case. Unsubscribe anytime.

Subscribed by employees and students at

Apple
Microsoft
LinkedIn
Columbia University
JioStar
Freshworks
Netflix
Databricks
MIT

".@DSPyOSS is so good that i'm kind of sad how many hours i spent struggling without it last year"

Joshua Weaver
Joshua Weaver
Director @txoji Attorney with a background in tech and entrepreneurism.

"Both DSPy and (especially) GEPA are currently severely under hyped in the AI context engineering world."

tobi lutke
Tobi Lutke
CEO by day, Dad in evening, hacker at night.

"We've reached that stage where every single day of the week (and every weekend), there are *several* really cool @DSPyOSS research papers, open-source applications, production use cases, or deep-dive tutorials, etc."

Omar Khattab

DSPyWeekly Issue No #21

Published on February 06, 2026

📚 Articles

Self-Optimizing Football Chatbot Guided by Domain Experts on Databricks

A practical guide to authoring, deploying, evaluating, and governing an agentic assistant that helps defensive coordinators anticipate opponent tendencies and continuously optimizes based on subject matter feedback.

How to Build Your Own Custom LLM Memory Layer from Scratch?.

Step-by-step guide to building autonomous memory retrieval systems

Evolving Node Prompts with GEPA on Open-Source LLMs

Now we face the challenge of improving prompting quality for our open models. We basically have two options: iterate manually or run some automation. This is where GEPA comes to our rescue.

DSPy: From Beginner to Advanced | atal upadhyay

A good code snippet driven style to teach DSPy.

🎥 Video

SpeechSage.ai : AI Receptionist powered by DSPy - YouTube

In this episode of the DSPY interview series I speaks with Noah Vandal, co-founder of Speech Sage, a company focused on integrating AI into healthcare communication. Noah shares his journey from aspiring doctor to electrical engineer, and how his background in computer vision and machine learning led him to develop AI solutions for healthcare. He discusses the challenges faced in the healthcare sector, particularly the inefficiencies and overburdened staff, and how Speech Sage aims to alleviate these issues by providing AI-driven receptionist services that enhance patient communication and appointment scheduling.

AI and Healthcare - Conversation with Dr. Marius Vach

In this episode of the Information Shelf - DSPy interview series, host Ankur Gupta speaks with Dr. Marius Vach, a board-certified radiologist and self-taught developer, about the intersection of medicine and technology. Dr. Vach shares his journey from medical school in Germany to becoming a medical AI researcher, highlighting his ability to bridge the gap between healthcare and coding. He discusses his creation of Wilhelm AI, an online QA system designed to make medical knowledge more accessible through AI, addressing the overwhelming amount of medical literature that radiologists must navigate daily. The conversation delves into the challenges of integrating AI into healthcare, the importance of domain knowledge in product development, and the potential for AI to automate repetitive tasks in radiology.

Planning, Reasoning, and Agents RG, 2026-01-14 Session: GEPA, prompt optimization outperforming RL. - YouTube

Dzmitry Pletnikau presents the paper "Reflective Prompt Evolution Can Outperform Reinforcement Learning" (GEPA). He situates the method within the DSPy ecosystem, contrasting prompt optimization with weight optimization techniques like GRPO by highlighting the former's ability to utilize rich natural language feedback. The talk details GEPA's evolutionary algorithm, which maintains a Pareto frontier of prompts to prevent mode collapse and enable cross-problem transfer. Dzmitry also compares the method to MIPROv2, noting that GEPA tends to produce more efficient, instruction-focused prompts rather than long few-shot examples.

🚀 Projects

GitHub - halfprice06/rlmgrep

Grep-shaped search and question answering powered by DSPy RLM. It accepts a natural-language query, scans the files you point at, and prints matching lines in a grep-like format. Use --answer to get a narrative response grounded in the selected files/directories.

dbreunig/dspy-monty-interpreter

Language: Python | License: MIT License

voicetestdev/voicetest

Test harness for voice agents. Import from Retell, VAPI, Bland, LiveKit. Run autonomous simulations. Evaluate with LLM judges. | Language: Python | License: Apache License 2.0

bojak83318/dspy-research

bojak83318/dspy-research

Jimmyu2foru18/github-analyzer

An AI-powered GitHub repository analyzer and build automation tool. This project helps developers analyze repositories, extract build instructions from READMEs, and automate build processes using OpenAI and DSPy. | Language: Python