Skip to content
VOL. 04 — JOURNAL OF THE NEAR FUTURE NO. 006

Currently — building agents in production

Notes from a solutions architect at the edge of cloud & AI.


Generative AICloud ArchitectureAgentic SystemsSecurityBedrockStrandsVoice RoboticsAWS Well-ArchitectedGenerative AICloud ArchitectureAgentic SystemsSecurityBedrockStrandsVoice RoboticsAWS Well-Architected

SECTION 01 / DISPATCHES

Latest dispatches

See full archive →

AI Engineering

The Agent Memory Problem: Why 5+ Solutions Exist and None Won

Mem0, Letta, Zep, graph-RAG, Neptune Memory, HiveMemory, Obsidian steering files -- the agent memory space is fragmenting faster than it's converging. Here's a landscape analysis of why no single solution wins, the four types of memory agents actually need, and a decision framework for choosing your architecture.

·10 MIN READ Read →
AI Engineering

MCP Gateway as Policy Enforcement Point: RBAC for Your Agent's Tool Access

Your AI agent has access to tools that perform real actions -- approving expenses, querying databases, modifying infrastructure. Prompt-based guardrails don't survive adversarial inputs. Here's how AgentCore Gateway + Cedar policies create a deterministic enforcement layer that operates independently of the agent's reasoning.

·9 MIN READ Read →

SECTION 02 / COLOPHON

The practice


FIELD

Generative AI, agentic systems, security architecture for regulated industries.

METHOD

Build it, measure it, write what worked. Anti-slop, anti-demo-ware.

STATIONS

AWS · ESIEA · CISSP · Lyon, France · Native FR / Pro EN, ES.

POSTAL

A monthly note, no algorithm.

One email a month — what I shipped, what surprised me, what I'd build if I had your weekend.

ONE LETTER A MONTH · NO TRACKER · UNSUBSCRIBE ANYTIME

END OF FOLIO

Have a project that won't fit in a tweet? Write to me.

Open a conversation

— TYPESET IN FRAUNCES & NEWSREADER —