... same knob, because they have very different costs. Scoring past 30 days is genuinely useless, for the reason you give. Storing past 30 days costs one hash per repo per scan, which is nothing. So keep the OID history unbounded and score only inside the window. A revert older than the window then produces a note rather than a finding: this exact content was here before, on this date. Not a security claim, just a fact the reader can weigh ... ... costs almost nothing to be able to say it. The reason I would push for unbounded

... rather than hold the line forever. A stalled job that took the cautious path is recoverable. A stalled job that just sat there is a support ticket. Comment from Dale Mooney(@dalemooney): @alex_amancio_candoa_   @aadilghani  Constant versus invariant is the useful thing here and I am taking it. One addition: you cannot separate them by observation, but you can by deliberately violating the constant once. A property that survives a test faking an unattended caller ... ... free. A human started session makes its promise at start time: somebody deliberately began something, and that is a commitment that it will either progress or end. So the expectation is not "this workspace should be busy", which is unbounded

... measure. Log the target URL, selected country, session identifier, HTTP status, and timing for every request. Treat target status codes separately from proxy connectivity errors mixing them up is a common debugging trap. 4. Set timeouts on every request Unbounded requests will hang your pipeline and burn through your bandwidth budget. Set reasonable timeouts (5 10 seconds is a good starting point) and implement retry logic with exponential backoff. A request that takes 30 seconds is almost certainly failing fail ...

... cohort is actually driving the bill, and it is almost never spread evenly. Once you can see it, the biggest customer side lever is usually not the model, it is the surface you give them. Free text in means unbounded context and unbounded

... Basedash s AI analyst is now more rigorous about the kinds of questions that trip up most tools. It s more careful with ambiguous metrics, distinguishes named rates from their supporting deltas, notices when a data source is stale versus fresh, and double-checks compound, multi-part answers before responding so you get answers you can trust, with the caveats that matter. We also taught it production-tested SQL gotchas for the databases you actually use Postgres, Supabase, Redshift, Snowflake ... ... usage Organization admins can now set an optional cap on AI overage spend right from billing settings. Once you reach your included credits plus the limit you ve set, AI usage pauses instead of running up an unbounded