Brian Clapper

I've authored a blog for developers who are new to vector embeddings and semantic search - if that's you, hopefully you'll find it helpful! #semanticsearch #vectorsearch #vectordatabase #vectorembeddings #densevectors #sparsevectors #elasticsearch https://lnkd.in/eAU8uACZ

36

1 Comment

Are you ready to say "Hasta la vista" to outdated search methods? Elastic is releasing its first live coding session: "How to Build Hybrid Search with Elasticsearch." What’s in it for you? ✅ Learn how to combine full-text search and semantic search into a powerful hybrid search solution ✅ Discover how retrievers like Reciprocal Rank Fusion (RRF) and Linear retrievers can fine-tune your search results ✅ Watch a live demo featuring a movie search app inspired by sci-fi classics Whether you're a developer, data scientist, or just passionate about AI-driven search technology, this session will equip you with the tools to build smarter, faster, and more relevant search engines. 📅 Date: July 21 📍 Location: YouTube 👉 Register here: https://gag.gl/LqiFi0

39

"I have only made this letter longer because I have not had the time to make it shorter." - Blaise Pascal'' This also applies to software demos. We've tightened the onboarding workflow and now have a 60 second demo of the journey from a blank slate in Claude Code to a deployed preview application on VeilStream.

1

When abstractions stabilize, code becomes habitable—and “LLM-friendly.” If your code uses small, intention-revealing helpers (e.g., write(), read(), partition(), delayWritesFrom(), and a HistoryRecorder that emits Jepsen EDN), you can prompt an LLM with paragraph from the book and get runnable tests for concrete scenarios. People reading patterns from my book or the DDIA book can try this with Tickloom. For example, see an interesting scenario from DDIA 10.6 section on 'linearizability and quorum' https://lnkd.in/dWMreGru Jepsen integration makes it concrete. #DistributedSystems #DDIA #Jepsen #LLM #Tickloom

26

1 Comment

During the "Ambient Agents" meetup last week, Brace Sproul gave us a walkthrough of Open SWE, a coding agent built on Langgraph. Open SWE monitors GitHub issues. When it gets a new one assigned, it kicks off a multi-agent architecture with a Manager, Planner, and Programmer. It can receive messages from users at any time during the process and incorporate them into the state. Check out the slides for diagrams explaining each agent: https://lnkd.in/g3PnNHqG And the code: https://lnkd.in/gY68TKYy I found the takeaways really interesting: * Summarization is hard: Especially with code changes that can get quite long, you literally can't send the entire context history to the LLM, and highly detailed old diffs may distract it. IIRC, the agent sends the last 20 messages, and summarizes the changes before. * Agentic > hard coded: Open SWE tends to use the agentic approach of "here are your tools, pick which one to use" for almost all tasks. Except for one.. * Explicit error handling: When an error happens, Open SWE stops everything and focuses on resolving the error. Otherwise, it could get too far afield and be unable to get back to a good place to solve the error. (Hopefully I didn't mis-represent anything I heard!) #Langchain #Langgraph

93

3 Comments

⚡ [ SYSTEM ALERT // JAN 11 ] 🛑 The most powerful coding agent is currently free. (But not for long). ———————————————————————— For the last year, we’ve been paying for "Copilots" and you hit the "Rate Limit" wall just when you get into the flow. As a Vision Architect, I am tracking a superior alternative: Cline. Enter Cline. It is not just a chat window; it is an autonomous engineer inside your VS Code. And right now, they are offering temporarily free access to premium inference models. This is an architectural window you need to exploit immediately. [ 🤖 COPILOT VS. CLINE (THE AGENTIC SHIFT) ] ———————————————————————— The difference isn't what they can do; it's how long they let you do it. GitHub Copilot (The Limited Assistant): The free tier allows LIMITED agentic interactions per month. Once you burn that fuel, you are downgraded to basic autocomplete. It is powerful but rationed. #Cline (The Autonomous Agent): Option A (Free): Use currently free/unlimited providers (like Grok or local Qwen via Ollama). Option B (Power): Plug in your own API key for Claude 3.5 Sonnet. You pay only for what you use, with zero artificial "rate limit" walls stopping your work. The Access: Currently, it provides free access windows to high-reasoning models (like Grok or devstral) without the aggressive throttling you see on Copilot. [ 🔓 THE ZERO-COST OPPORTUNITY ] ———————————————————————— Usually, "Agentic" workflows burn through API tokens instantly because the AI has to "think" in loops. It gets expensive fast. The Current Protocol: #Cline is currently providing free access to select high-performance models. This removes the "Token Tax" from experimentation. You can run complex, multi-step refactoring tasks that would normally cost $5–$10 in API credits for absolutely zero cost. [ 🛠️ ARCHITECTURAL USE CASE ] ———————————————————————— Don't use this for "Hello World." Use the free window for heavy lifting: Legacy Refactor: "Read the entire /src folder and migrate all class components to functional components." Test Generation: "Analyze my API routes and generate a full Jest test suite. Run the tests and fix any failures." [ 🔓 WHY SWITCH NOW? ] ———————————————————————— Copilot: "You have used your credits. Please upgrade." 🛑 Cline: "I fixed the bug. What would you like me to build next?" 🚀 [ 🎯 THE ARCHITECT’S VERDICT ] ———————————————————————— The window to test "Agentic Workflows" for free is closing. We are moving from "Human-in-the-Loop" to "Human-on-the-Loop." Download #Cline from https://cline.bot, and let it handle the grunt work while you focus on the architecture. When the "Free" meter runs out on Copilot, don't reach for your wallet. Reach for a better architecture. STATUS: 🟢  Unrestricted Agent Active. ACTION: ⚡Deploy Cline + Devstral (Free). Q: Have you hit the "Rate Limit" wall on #Copilot yet? 👇 [ METADATA ] Author: Dhruvkumar V Role: #VisionArchitect Focus: #Cline #AI #AgenticWorkflows #VSCode #CodingAgent

2