Skip to content
~/sph.sh

python

6 posts

python
HMAC: What It Is, How It Works, and When It Is the Wrong Tool

HMAC-SHA-256 for webhooks, signed URLs, and internal auth, with runnable code in three languages and the boundary where digital signatures take over.

security · encryption · webhooks +4

MCP vs Direct API Access for AI Agents: When to Skip the MCP Layer

Why production teams replace broad MCP access with scoped API proxies. Atlassian, Google Workspace, and Notion via FastAPI proxy, CLI wrapper, and n8n.

mcp · api-design · python +4

Prompt Engineering for Production Systems: A Systematic Engineering Approach

A technical guide to production-grade prompt engineering: systematic design, security, observability, and cost optimization for enterprise LLM apps.

prompt-engineering · llm · ai-tools +6

AI Agent Security: Guardrails and Defense Patterns for Production Systems

Securing AI agents in production with AWS Bedrock Guardrails, defense-in-depth, and patterns that prevent prompt injection, tool misuse, and multi-agent attacks.

ai-agents · aws-bedrock · security +3

LangChain in Production: Patterns That Work and Anti-Patterns That Don't

Lessons from running LangChain in production: the anti-patterns that cause failures, the patterns that work, with code examples and cost optimization strategies.

langchain · llm · production +5

OpenTelemetry Fundamentals: A Beginner's Guide to Modern Observability

A beginner's guide to OpenTelemetry covering traces, metrics, and logs with practical implementation examples, common pitfalls, and a terminology glossary.

opentelemetry · observability · monitoring +4