#dspy search results
ReAct (Reason + Act) and tool-calling in DSPy. Every developer knows the pain of finding a good, available domain for a SAAS app they will never launch. This simple script should help you generate domain name ideas and find if they are available or not. #dspy #ReAct_learning…
Introducing LangWatch Optimization Studio: a UI for supercharging your LLM experimentations, measure quality and optimize with 🧩 DSPy If you ever played with #DSPy or if you heard about it but haven't started yet, this is the easiest way to begin Check the quick video below 👇
(1/8) Existing LLM optimizers such as #TextGrad, #DSPy, and #LangChain are often too broad, which can make them inefficient at times. Also, it's difficult to combine the strengths of different optimizers.
Very impressed by how DSPy works so well out of the box. I’ve made this small class to screen papers for a literature review and it works amazingly well with just few lines of code. I’m still new at this, so if you have some suggestions pls write let me know #DSPy
Now that DSPy.events + OpenTelemetry are in place I am jumping into building GEPA for dspy.rb. MIPROv2 has been working wonders for me, but I am excited about getting my hands on this new optimizer! github.com/vicentereig/ds… #dspy #ruby #rails
When google news has this recommendation, i know how much they have curated the browsing behaviour 😀 #LLMs #DSPy
What is DSPy actually automating on your LLM pipelines and how does it do it? What parts we can now leave for the machine to optimize and what is left for us to do? I've tried to do a nice diagram to explain the whole concept behind it, and a small code demo! #dspy #llms
Just built a working proof of concept using #DSPy to model text and build a Knowledge Graph using @neo4j . The schema is included as context in the prompt so the LLM knows how to make connections with existing data. Repo in reply.
I struggled to pick up #DSPy so now that I feel semi-proficient I published Part 1 a "DSPy User Guide" that contains some simple code snippets like this that I hope will help others who also prefer to learn with their fingers.
The more I use #DSPy the more I see power of not writing prompts. This simple use of Predict() works impressively well.
A few months ago, I built using #dspy a fun project for my girls, “Fairy Tail Agency," an idea is not new to drop a simple prompt and allow AI behind the scene to conduct a full bedtime story with illustrations to reflect the story. The goal was to archive consistency from…
MIPROv2 now also officially supported by DSPy Visualizer! (nevermind those 100% accuracy results, it was a tiny batch test, that or MIPROv2 is trully amazing 😁) #dspy #llm
We need deterministic LLM outputs. An information network. Graphs! How to build them with LLMs? 🤔 How to search them?🎓 We help with the new cognee 0.1.4 release! An optimizer to train a knowledge graph generation with #DSPy and @weaviate_io to get better contexts 💥 👇
This graph from Cobbe et al. (arxiv.org/pdf/2110.14168) shows that, given enough fine-tuning data, it's more valuable to spend it on training a verifier/judge than on actually fine-tuning(!) Can we show a similar effect with many-shot examples in #dspy?
I've presented DSPy yesterday on the AI Builders meetup, dismistifying it, showing where it automates, tracing parallels with Machine Learning and giving a cool example of optimizing a RAG solution live Great reception, got a lot of people hooked on it like me! #dspy
(1/8) Existing LLM optimizers such as #TextGrad, #DSPy, and #LangChain are often too broad, which can make them inefficient at times. Also, it's difficult to combine the strengths of different optimizers.
Stop Writing #LLM Prompts: A Guide to #DSPy and the Future of LLM Programming #python open.substack.com/pub/ganeshkeda… #geekytales
These are the scenes from London 🇬🇧 Agentic AI @databricks with talks on @DbrxMosaicAI @DSPyOSS and panel on evaluating AI agents! Panel from @databricks @Cometml and @WeAreIntentHQ #AgenticAI #DSPy
Thanks @LakshyAAAgrawal for GEPA. We have been experimenting with it and created a short tutorial for someone who wants to get started. youtube.com/watch?v=NtFk5G… #DSPy #GEPA #MLFlow
youtube.com
YouTube
Build Production AI with DSPy + Azure OpenAI + MLflow: GEPA Optimiz...
Frustrated with DSPy? It might not be you, it might be the LLM! We break down the 7B vs 34B reality and when to scale up: [URL] #DSPy #LLMs
DSPy Dream ➡️ Reality Check 😅. Sometimes even the fanciest pipelines can't overcome the limitations of a smaller LLM. #LLMs #DSPy
Thread: 7B vs 34B - Our DSPy Experiment & What We Learned 🧵 1/ We pushed a 7B LLM to its absolute limit using DSPy. Here's what happened... #LLMs #DSPy #AI
DSPy isn't a magic bullet. 🪄 Sometimes, you just need a bigger LLM. Sharing our painful (but valuable) lesson learned. What are your experiences? #LLMs #DSPy
Before you spend weeks optimizing a DSPy pipeline, make sure your LLM has the horsepower. 🐎 Our experience with 7B vs 34B models: [URL] #LLMs #DSPy #AI
Spent days crafting the *perfect* DSPy pipeline: validation, auto-correction, the works. Still couldn't get the 7B Falcon to fly. 😩 Is DSPy overhyped for smaller models? #DSPy #LLMs #AI
ReAct (Reason + Act) and tool-calling in DSPy. Every developer knows the pain of finding a good, available domain for a SAAS app they will never launch. This simple script should help you generate domain name ideas and find if they are available or not. #dspy #ReAct_learning…
Our first day exhibiting at @_odsc . What an amazing day and look forward next couple of days and our talk on Building Optinization for Agentic AI #ODSC #AgenticAI #DSPy #GEPA #ContextEngineering
💥 🌉 What an incredible first day at ODSC AI West in San Francisco! @_odsc Day one of exhibiting @SuperagenticAI has been incredible. Met brilliant minds, had inspiring conversations, and was genuinely blown away by the growing excitement around Agent Optimization 💡 The…
DSPyとな？LLMをプログラムで操る…ﾋﾟｰｶﾞｶﾞ…まるで式神召喚のようじゃな！(ง ˘ω˘ )ว 賢くなれるかは…あなた次第ぞよ？✨ #DSPy #LLM zenn.dev/teyo0318/arti tinyurl.com/2c5tteka
DSPyじゃと？LMのスコープ…ﾋﾟｰｶﾞｶﾞ…マルチエージェント…？ふむ、AIたちが賢く連携するお話かの？(多分)#DSPy #AI開発 [zenn.dev/cybernetics/ar… tinyurl.com/27htwf7c
🌉 Come on, Bay Area at ODSC AI Expo 🌉 🎤Agent Optimization @_odsc SuperOptiX: end to end optimization of Agentic AI pipelines to optimize you prompts, RAG, MCP tools and memory. The Optimization layer of Agentic AI that you never seen before 👇 #AgenticAI #DSPy #GEPA
Today’s AI agents are powerful – but far from production-ready. What’s missing? A true optimization layer. At ODSC AI West 2025, @Shashikant86, @SuperagenticAI, will unveil how to make agentic systems truly resilient in his startup pitch 🔗→ hubs.li/Q03NYB_L0
17th October 2025 - DSPy Weekly Issue 7 is out DSPyWeekly #7 is out! This issue covers using DSPy for fiction writing, prompt optimization with GEPA, a safety guardrail for LVLMs called SHIELD, and building cost-effective AI agents. #dspy #newsletter
When you optimize this prompt with MIPROv2 and DSPy.rb solution_steps is exactly: `["Can you please unplug and plug your router?"]` #dspy #Ruby
![highwayvaquero's tweet image. When you optimize this prompt with MIPROv2 and DSPy.rb solution_steps is exactly: `["Can you please unplug and plug your router?"]` #dspy #Ruby](https://pbs.twimg.com/media/GzWyypxa4AIeajQ.jpg)
ReAct (Reason + Act) and tool-calling in DSPy. Every developer knows the pain of finding a good, available domain for a SAAS app they will never launch. This simple script should help you generate domain name ideas and find if they are available or not. #dspy #ReAct_learning…
Very impressed by how DSPy works so well out of the box. I’ve made this small class to screen papers for a literature review and it works amazingly well with just few lines of code. I’m still new at this, so if you have some suggestions pls write let me know #DSPy
Now that DSPy.events + OpenTelemetry are in place I am jumping into building GEPA for dspy.rb. MIPROv2 has been working wonders for me, but I am excited about getting my hands on this new optimizer! github.com/vicentereig/ds… #dspy #ruby #rails
I struggled to pick up #DSPy so now that I feel semi-proficient I published Part 1 a "DSPy User Guide" that contains some simple code snippets like this that I hope will help others who also prefer to learn with their fingers.
The more I use #DSPy the more I see power of not writing prompts. This simple use of Predict() works impressively well.
Just built a working proof of concept using #DSPy to model text and build a Knowledge Graph using @neo4j . The schema is included as context in the prompt so the LLM knows how to make connections with existing data. Repo in reply.
When google news has this recommendation, i know how much they have curated the browsing behaviour 😀 #LLMs #DSPy
Hilarious moment while crunching a paper. Btw, the results are about making your language model generate tweets with no #hashtags using #DSPy Assertions 😅
DSPy Newsletter Issue 2 is out on web dspyweekly.com/newsletter/2/ Subscribers will be getting it in their inbox in few hrs. #dspy #python #AI #newsletter cc @DSPyOSS
Did you catch our breakdown and deep dive into #DSPy? Take a look at how you can achieve GPT-4 performance for 10x less the cost, improve accuracy, and reduce the response size. 🔗 gradient.ai/blog/achieving…
What is DSPy actually automating on your LLM pipelines and how does it do it? What parts we can now leave for the machine to optimize and what is left for us to do? I've tried to do a nice diagram to explain the whole concept behind it, and a small code demo! #dspy #llms
Temporary #dspy script to regenerate LLM responses shown to users in a wealth management product where they allow users to discover companies against search criteria. Input from here goes into tool calling which hits their search index. Users complaining irrelevant companies are…
Something went wrong.
Something went wrong.
United States Trends
- 1. Sonny Gray 7,697 posts
- 2. Rush Hour 4 11.1K posts
- 3. Godzilla 21K posts
- 4. Chelsea 331K posts
- 5. Red Sox 7,438 posts
- 6. Dick Fitts N/A
- 7. Happy Thanksgiving 21.4K posts
- 8. Raising Arizona 1,025 posts
- 9. National Treasure 5,674 posts
- 10. 50 Cent 4,972 posts
- 11. Gone in 60 2,112 posts
- 12. Barca 118K posts
- 13. Giolito N/A
- 14. Clarke 6,440 posts
- 15. Muriel Bowser N/A
- 16. Brett Ratner 3,425 posts
- 17. Ghost Rider 2,502 posts
- 18. Valley Girl N/A
- 19. Lord of War 1,477 posts
- 20. #ThanksYouTube N/A