Kelet – Root Cause Analysis agent for your LLM apps

https://kelet.ai/

#HackerNews #Kelet #RootCauseAnalysis #LLMapps #AItechnology #SoftwareDevelopment

Just shipped a deep-dive for Java devs who want real stateful LLM apps.
Persistent memory. Sliding windows. Multi-session isolation.
All powered by Quarkus + LangChain4j + JPA.

If you want your AI assistant to remember across pods and restarts, this tutorial will get you there.

🔗 https://www.the-main-thread.com/p/persistent-llm-memory-java-quarkus-langchain4j-tutorial

#Java #Quarkus #LangChain4j #AIEngineering #LLMApps #OpenSource

AI apps monetization is 🔥 but tricky.

What's your biggest blocker today?

A) Not sure which prompts have value
B) Don't want to ruin user experience
C) No privacy-safe solution
D) Can’t track intent clusters

Like + reply 👇
(We’re building VexRail to solve all 4)
#AI #LLMapps #StartupDev

Exciting news! LangSmith now offers dynamic few-shot example selectors, allowing you to rapidly improve LLM app performance with just one click. Say goodbye to manual dataset curation and hello to efficiency! 🚀 #LangSmith #LLMApps

https://blog.langchain.dev/dynamic-few-shot-examples-langsmith-datasets/

𝐂𝐨𝐩𝐢𝐥𝐨𝐭 𝐟𝐨𝐫 𝐒𝐞𝐜𝐮𝐫𝐢𝐭𝐲: 𝐞𝐥𝐞𝐦𝐞𝐧𝐭𝐬 𝐨𝐟 𝐚𝐧 𝐞𝐟𝐟𝐞𝐜𝐭𝐢𝐯𝐞 𝐩𝐫𝐨𝐦𝐩𝐭

From the "Get started with Microsoft Copilot for Security" online training, I highlight this interesting in-depth analysis.

𝐄𝐟𝐟𝐞𝐜𝐭𝐢𝐯𝐞 𝐩𝐫𝐨𝐦𝐩𝐭𝐬 give Copilot adequate and useful parameters to generate a valuable response. Security analysts or researchers should include the following elements when writing a prompt.

💡 𝐆𝐨𝐚𝐥 - specific, security-related information that you need

💡𝐂𝐨𝐧𝐭𝐞𝐱𝐭 - why you need this information or how you'll use it

💡𝐄𝐱𝐩𝐞𝐜𝐭𝐚𝐭𝐢𝐨𝐧𝐬 - format or target audience you want the response tailored to

💡𝐒𝐨𝐮𝐫𝐜𝐞 - known information, data sources, or plugins Copilot should use

At this link other prompting tips:

https://learn.microsoft.com/en-us/training/modules/security-copilot-getting-started/5-create-effective-prompts

Full training: https://learn.microsoft.com/en-us/training/paths/security-copilot-and-ai/

#copilot #copilotforsecurity #securitycopilot #microsoft #microosoftsecurity #llm #openai #azureopenai #llmapps #soc #generativeai #genai #cybersecurity #azure #cloudsecurity #cloudnative #defender #sentinel #microsoftsentinel #xdr #defenderxdr #prompt #promptengineering