SSharon Langer·6d agoSmall Models, Big Agents: Are We Teaching the Wrong Model?System One, System Two, and a different way of thinking about AI agents.9 min read·sharonlanger.hashnode.dev11A
DDataNorthAI·6d agoSmall Language Models vs LLMs: Key DifferencesCompare small language models and LLMs on cost, speed, deployment and specialisation, with practical business use cases.4 min read·datanorth.hashnode.dev
SSyncfusion, Inc.·Nov 3Jev AI vs. Structured-Output LLMs: Which One Should You Actually Use?Jev AI vs. structured-output LLMs: understand the differences, use cases, and how clean document data improves reliable AI decisions.19 min read·syncfusion-blogs.hashnode.dev
AAhmed Shendy·Nov 2Blackboard Architecture with LLM Agents: What Could It Make Possible?Exploring how LLM agents use Blackboard Architecture to discover data and coordinate through shared request and response boards without direct task assignment.3 min read·shendy.hashnode.dev
AAdam Jaber·Oct 30Why AI Coding Assistants Suggest Deprecated APIs (and How to Stop It)It's not just the knowledge cutoff. The old API has more votes in the training data, and your own code decides which era the model writes in.9 min read·simplyexplained.hashnode.dev
NNash Hasan·Oct 29Closing the Prompt Injection Vulnerability: Why Heads of Security Changes Enterprise Agent Risk for Higher EducationLearned attention gates drop prompt injection attack rates to 10 percent without breaking administrative tool calling, establishing a verifiable baseline for enterprise AI governance.8 min read·blog.edtechniti.com
NNash Hasan·Oct 23The Inference Wall Hits Campus IT: Direct Pass@k Optimization and the Economics of Test-Time ComputeEarly thought pruning cuts multi-step reasoning compute by over 60% while maintaining solution diversity and institutional compliance.7 min read·blog.edtechniti.com
AAhmed Adawy·Oct 22Your LLM Is Slow. The GPU Might Not Be the Problem.Learn how to diagnose slow LLM APIs by separating queueing, retrieval, prefill, decoding, and streaming latency before changing hardware. Python soft9 min read·adawy.hashnode.dev
AAdam Jaber·Oct 22RAG Chunking: Why Chunk Size Is the Wrong KnobMost wrong RAG answers come from chunks that lost their context, not chunks of the wrong size. Here's how to find them and fix them.9 min read·simplyexplained.hashnode.dev
NNash Hasan·Oct 20Deterministic JSON Reaches Claude on Bedrock: Why Constrained Decoding Changes Campus AI InfrastructureNative constrained decoding on AWS Bedrock eliminates JSON schema failures, stabilizing campus AI integration and API costs.7 min read·blog.edtechniti.com1A
