Let an AI sort the inbox, then snooze what can wait
How a small AI model labels each new mail as high or low value, why low-value mail is snoozed to two fixed times a day, and why I let it err towards important.
Reaching maximum efficiency wherever automation is feasible: workflows, scripts and tools that take repetitive work off people's hands.
Automation starts with an honest look at the work: which tasks come back every day, week or month and follow the same steps each time. That is where software beats people, not because it is clever, but because it never gets bored and never skips a step. The hard part is rarely the automation itself. It is describing the process clearly enough.
I spent more than twelve years in ICT management in automotive and manufacturing before going full-time on my own businesses in 2021, and the drive has stayed the same: maximum efficiency wherever automation is feasible. Tools like n8n, Python and SQL play big roles in that. AI has added a new layer, because steps that used to need a person reading text can now often be handled by a language model.
Outside work I tinker with home automation, which teaches the same lessons on a smaller scale. An automation you cannot see is one you cannot trust, so every workflow should report clearly when it fails, and a failure should never look like a quiet zero.
Longer articles about Automation
How a small AI model labels each new mail as high or low value, why low-value mail is snoozed to two fixed times a day, and why I let it err towards important.
Blockspot, VoucherAlarm and Automate & Tweak can now post to the updates feed on this site through its own API, with the guardrails in the API instead of a review step.
About Automation, from Blockspot only
How a small AI model labels each new mail as high or low value, why low-value mail is snoozed to two fixed times a day, and why I let it err towards important.