October 12, 2026 · Reqmint
Background check hits arrive on a timer. This walkthrough builds the state machine, tables and daily exceptions list behind a pre-adverse action workflow: who gets what document, how long the waiting period runs, where the individualized assessment is recorded, and the three steps the agent never takes on its own.
Read moreOctober 5, 2026 · Reqmint
Run the agent read-only on real data, log what it would have done, and label a sample. This walkthrough builds the shadow table, the backtest that does not leak, the weekly recruiter review, and the four numbers that decide go, narrow or not yet.
Read moreSeptember 28, 2026 · Reqmint
Pay transparency rules now bite in Illinois, Minnesota, New Jersey, Vermont and Massachusetts, and staffing firms post on behalf of clients. Here is a rules table, a jurisdiction resolver for remote and multi-state reqs, and a pre-publish gate that keeps a record of what was posted where.
Read moreSeptember 21, 2026 · Reqmint
MSP and VMS programs distribute reqs by email with deadlines measured in hours. A walkthrough of parsing them into a req table, handling time zones and dedupe, and letting a person create the ATS job.
Read moreSeptember 21, 2026 · Reqmint
Submission-to-interview time is mostly waiting for three calendars to agree. A walkthrough of the scheduling agent that fixes it: free/busy reads, a holds table with a TTL, timezone rules, and the reschedule path everyone forgets to build.
Read moreSeptember 18, 2026 · Reqmint
Back-office agents rarely fail loudly. They fail by going quiet: a rotated token, a renamed field, a paused schedule. Here is the monitoring layer we build alongside every agent — a run log, per-stage counters, staleness checks on the work itself, and a digest a named human reads.
Read moreSeptember 17, 2026 · Reqmint
Hours worked against an exhausted purchase order are a negotiation, not revenue. This walkthrough builds a burn-down agent: a PO table with real caps, consumption in three buckets, a burn-rate projection, four rules, and a drafted extension request a human sends.
Read moreSeptember 15, 2026 · Reqmint
Model and API spend arrives as one monthly invoice with no idea which workflow spent it. This walkthrough builds the run ledger: one row per run, cost captured at the step, cost per outcome rather than per token, a cap that stops a runaway loop, and the monthly query that tells you which agent to switch off.
Read moreSeptember 14, 2026 · Reqmint
jobs@ and info@ are where reqs, resumes and timesheet chases arrive together. A walkthrough of a triage agent that reads a shared mailbox, classifies each thread, deduplicates it against the ATS, and hands a recruiter a one-click action instead of a folder.
Read moreSeptember 14, 2026 · Reqmint
Before an agent drafts anything it needs credentials. This walkthrough builds the access layer: one service identity per agent, a narrow tool surface instead of raw API keys, read and write split, and an access log that answers who touched which candidate record and why.
Read moreSeptember 11, 2026 · Reqmint
Cross-system agents stall on identity, not models: the same contractor is three spellings and four ids across ATS, timesheets, payroll and the inbox. This walkthrough builds the resolution layer: a person and link table, normalisation, deterministic rules first, a scored middle band with a review queue, and merges you can undo.
Read moreSeptember 9, 2026 · Reqmint
An agent is only as current as its copy of the ATS. This walkthrough builds the sync layer underneath one: an event ledger, webhook signature checks, deduplication, a nightly catch-up pull, and a staleness number you can put on a dashboard.
Read moreSeptember 9, 2026 · Reqmint
Logging every automated screening decision is the easy half. This walkthrough builds the other half: a retention table with a clock per record class, partitioned decision logs, legal holds that beat the purge job, and a purge that leaves proof it ran.
Read moreSeptember 9, 2026 · Reqmint
A screening agent tested once, by hand, drifts the moment a model version or a prompt changes. This walkthrough builds the harness: a frozen golden set, pinned model and rubric versions, an agreement and movement diff, and rationale checks that fail the run.
Read moreSeptember 9, 2026 · Reqmint
Approved hours, the pay run and the invoice run drift apart every cycle, and the difference is margin. This walkthrough builds the reconciliation: three landing tables, a rate card with effective dates, five exception rules in SQL, and a Tuesday exceptions list a human works.
Read moreSeptember 9, 2026 · Reqmint
A finishing-soon list tells you who to call, not what about. This walkthrough builds the matching step: a skill alias table, hard filters before scoring, a four-component score with tunable weights, stored rationale per pair, and the narrow job a model should actually do.
Read moreSeptember 9, 2026 · Reqmint
An agent that drafts recruiter outreach needs somewhere for a human to say yes. This walkthrough builds the queue: a suppression check before drafting, one row per draft with a state machine, a claim step so two recruiters cannot send the same message, expiry for stale drafts, and the write-back that puts the sent note on the ATS record.
Read moreAugust 22, 2026 · Reqmint
A hands-on walkthrough of getting reliable contractor end dates out of an ATS API: OAuth 2.0 setup, paginated placement pulls, an idempotent landing table, and the reconciliation step most pipelines skip.
Read moreAugust 15, 2026 · Reqmint
Why firms lose their finishing contractors, how to measure the leak, and what an agent needs to close it.
Read moreJuly 18, 2026 · Reqmint
Most first agents fail at scoping, not engineering: a worked example using the Monday-morning timesheet chase.
Read more