Deep dives into artificial intelligence, large language models, transformers, embeddings, and the core concepts powering modern AI systems.
25 articles in this seriesBoulder uses 9 Strands agents on Bedrock AgentCore to generate, deploy, and maintain full-stack apps on AWS Amplify â with self-healing builds and self-improving prompts.
Latest Post
A discovery call with a global specialty chemicals company revealed that the real AI bottleneck isn't models â it's data. Here's what enterprise chemistry teams actually need versus what the hype promises.
Strands Agents plugins let you intercept every decision in the agentic loop. Steering hooks achieved 100% accuracy across 600 evaluation runs â where prompt engineering scored 82.5% and graph workflows 80.8%.
Most RAG tutorials stop at 'put vectors in a database.' This post covers what actually determines quality: how you chunk documents, which vector search engine to pick, and how to measure and iterate on retrieval performance using Bedrock Knowledge Bases and LLM-as-judge evaluation.
Vector search, semantic search, keyword search, hybrid search â these terms get used interchangeably but they mean different things. This post breaks down what each actually does, when each matters, and why hybrid search wins for RAG.
A deep architectural comparison of four open-source frameworks that turn messaging apps into AI assistant interfaces â from a 349-file TypeScript monolith to a 10MB Go binary that runs on a $10 board.
A deep dive into the multi-agent architecture behind AWS Security Agent's automated penetration testing â from specialized agent swarms to assertion-based validation.
A 5-component framework for writing effective system prompts for any AI agent â Bedrock Agents, Claude Code, LangChain, Strands, or custom builds. With a practical Claude Code implementation.
A deep dive into World Monitor â an open-source intelligence dashboard that aggregates 150+ feeds, 40+ geospatial layers, and AI-powered analysis into a real-time situational awareness platform. What OSINT is, how these platforms work under the hood, and why it matters now more than ever.
A beginner-friendly walkthrough of how an LLM actually works end-to-end: from typing a prompt to receiving a response â covering tokenization, embeddings, Transformer layers, KV cache, the training loop, embeddings for search, and why decoder-only models won.
The 5 key concepts every cloud architect should know about LLM serving: PagedAttention, KV cache mechanics, continuous batching, MoE trade-offs, and real production numbers.
Turn Claude Code into a persistent executive assistant with morning briefings, auto-logging, context-aware reminders, complex skills, and a memory that compounds over time â using only markdown files.
Two strategies to shrink LLMs â one compresses weights, the other transfers knowledge. A practical guide to distillation and quantization: when to use each, how to implement them with Hugging Face, and why the real answer is both.
What TFLOPS actually measures, why FP16 matters for LLMs, and why the most important GPU bottleneck for inference isn't compute at all.
A practical walkthrough of two paths to working with Mistral â the managed API for fast prototyping and self-hosted deployment for full control â with real code covering prompting, model selection, function calling, RAG, and INT8 quantization.
Everything a cloud/AWS engineer needs to know about Python, the Hugging Face Transformers framework, SageMaker integration, quantization, CUDA, and AWS Inferentia â without being a data scientist.
A deep dive into the Transformer architecture â how attention connects tokens and why the Feed-Forward Network is the real brain of the model. Plus the key to understanding Mixture of Experts (MoE).
End-to-end guide: fine-tune Mistral models with LoRA using Hugging Face Transformers, then deploy at scale with vLLM on AWS â from training to production serving on SageMaker, ECS, or Bedrock.
Unitree ships a humanoid robot with 43 degrees of freedom, a full AI training pipeline on GitHub, and Apple Vision Pro teleoperation â for $13.5K. Here's what the developer ecosystem looks like.
AWS now offers 9 different ways to store and search vectors for RAG workloads. This guide compares every option through the Well-Architected Framework to help you pick the right one.
A practical walkthrough of how large language models are aligned with human values â from collecting feedback to PPO optimization and the reward hacking pitfalls.
Understand how LLM fine-tuning works, when to use it, and how to choose between full fine-tuning, LoRA, soft prompts, and other PEFT methods.
AI platform teams need governance before scaling. Learn how to use Amazon Bedrock inference profiles, AWS Budgets, and a proactive cost control pattern to track, allocate, and cap AI spending per team.
Discover AI-DLC (AI Development Lifecycle), a structured framework for AI-assisted software development. Learn how I used it to build this blog from scratch and how it enables continuous iteration.
A practical guide to building generative AI applications with Amazon Bedrock
Best practices and deep dives into cloud security â from zero trust architectures to WAF configurations and penetration testing.
15 articlesPatterns, services, and architectural decisions for building on AWS â including Bedrock, SageMaker, Transfer Family, and more.
39 articlesExploring the frontier of autonomous AI agents â multi-agent systems, RAG pipelines, and agentic frameworks like Strands.
11 articles