Hi, I'm Farah.
Tech talk unleashed — articles, research notes, and architecture discussions on distributed systems, AI agent pipelines, and cloud infrastructure.
Last updated: September 2026
Working on
- Agent governance and containment patterns for production systems.
- Writing up the pieces of my career story I haven’t told publicly yet.
Reading / thinking about
- How eval harnesses fail when the sandbox leaks into the real internet.
- Where local-first, privacy-respecting AI actually beats the cloud-default.
Elsewhere
Still posting on X, still pushing to GitHub.
This is a now page, inspired by Derek Sivers.
GitHub
Farah (@The-Farah)
Farah (@farahInTheZone)Browse by Topic
Recent Writing
Session, Store or Weights: Where Should Your Agent Remember Things?
The brain stores and connects memories, and an agent stack does the same in three places: the session, a store it can search, and the weights. A plain guide to which fact belongs where, and what to let it forget.
Where Does Turn 12 Go? Load Balancing Agent Sessions
Scalable Backends, stop 2: L4 vs L7, round-robin vs consistent hashing, and why sticky sessions fight you, told through AgentDesk's long-running agent sessions.
No One Starts Again After 13 Years. I Did.
What I relearned first after thirteen years away from tech, and what hadn't changed at all.
"Does It Scale?" Is Three Questions Pretending to Be One
Scalable Backends, stop 1: latency, throughput and availability, why they trade against each other, and why agent backends feel the trade first. Told through AgentDesk's overnight jump from 100 to 100,000 sessions.
Direction vs. Destination: When a Vague Prompt Is Actually Fine
Four real prompts that fixed my profile picture, and the rule they taught me: match how precise you are to how long you'll wait to find out you were wrong.
AgentDesk: The System Behind the Scalable Backends Series
Meet AgentDesk, the fictional AI agent platform used as the running example throughout the Scalable Backends teaching series — and its baseline container architecture.