sema.cloud
Writing

Notes from the build.

Architecture decisions, benchmark write-ups, and the occasional retrospective on what I got wrong. Every post is grounded in a real project on this site.

Posts

All writing

AI Security

The Security Boundary Is Not The Prompt

Multi-tenant AI systems should enforce tenant scope before retrieval. Tenant Lens uses tenant-scoped ingestion, per-tenant indexes, and authorization before data access.

Read post
AI Engineering

Shiny AI Demos Need Promotion Gates

A working AI demo is not the same thing as a promotable workflow. AgentOps Lab explores quality, trace, latency, and cost gates for RAG and agentic systems.

Read post
GenAI / RAG

RAG Retrieval Tuning

Tenant Lens improved from 48.57% to 94.29% by treating retrieval as an engineering problem, not by swapping the language model.

Read post
AWS / Platform

Honestly, How Long Does It Really Take to Provision an AWS Landing Zone?

Control Tower can start an AWS landing zone in under two hours. A production-ready landing zone with identity, logging, security, networking, automation, and account vending usually takes weeks.

Read post
Cloud Native

A Model Deprecation Validated My Platform Design

Anthropic deprecated Claude 3 Haiku on Bedrock mid-benchmark. The reusable challenger lane absorbed the full swap to Nova Lite as a config change. The replacement was 28× cheaper and 6× faster than Textract.

Read post
ML / RAG

Building a Multi-Tenant RAG Platform on AWS

Tenant Lens is a production-shaped RAG MVP with tenant isolation, reliable ingestion, authorization before retrieval, hybrid search, grounded answers, and measured retrieval quality.

Read post