Urgent.News

What's breaking now, across thousands of outlets.

Tech

The voice that says "not ready" is older than your new job

Under a post of mine about portfolios, someone wrote that a README proves the work and never records what it was like before the work existed. Under another, someone described watching their own project stop at eighty per cent, and said the reason was stupid when they finally found it: the doubt that kept them going, the doubt that they could finish it, had been answered. Two different people.…

The hour-long exercise focuses on helping individuals confront their doubts and fears about starting a new career or project. It emphasizes that the doubts are not tied to the new job itself, but rather to an older, ingrained voice of uncertainty. The process involves identifying this voice, understanding its origins, and taking a small, reversible action within the next 72 hours to build confidence.

The exercise encourages participants to avoid over-planning and to act despite fear, as the real danger lies in the fear being visible to others. The ultimate goal is to identify a single, manageable step to take and to share it openly with someone, rather than hiding behind preparations.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in Tech

Idempotency Keys and Distributed Transactions Prevent Double Payouts During Network Timeouts

The Anatomy of a Cross-Border Phantom Payout When an application client initiates a cross-border transfer, it projects an illusion of atomicity—a single, monolithic API request that cleanly moves…

  • Double payouts during network timeouts stem from distributed transaction complexity.
  • Idempotency keys with status inquiries prevent duplicate payouts in transient network partitions.
  • Circuit breakers are needed when partner systems are down or inconsistent.

Run your own bake-off before you pick a production model

Public leaderboards will not tell you which model should run your highest-volume prompt. I'd start with one narrow step you can score, a labelled set built from real production misses, and a side-by-side run against a frontier model plus one small language model candidate.

  • Run a narrow bake-off with labelled production misses
  • Compare frontier model and small language model side-by-side
  • Choose candidate with stable high-volume ancillary work

More from Saturday 14 November →