QC releaseyour reviewer sees only what failed.
By the time a batch reaches QC, every signed step has already rolled up. Your reviewer reviews-by-exception, signs, and releases — with the deviations and CAPAs already linked.
What changes once QC review-by-exception & batch release is live.
Indicative ranges from V5 pilot deployments. Your numbers will land near these once the workflow is operator-led and e-signed at the step.
What changes the day you switch this on.
QA reviews every batch line-by-line — release queue keeps growing.
Manual, paper-driven, and only audited after the fact.
Tribal knowledge in spreadsheets and shared drives.
Wet-ink signatures collected hours after the work happened.
Review-by-exception queue
Deviation + CAPA built in
CoA from real data
Two-signature release
What you actually get
Operator-led, e-signed, immutable. Engineered for regulated manufacturers — not retrofitted.
Review-by-exception queue
Pass-pass-pass batches are one click. Anything out-of-spec is highlighted with the failing data point.
Deviation + CAPA built in
Out-of-tolerance events open structured deviations linked to operator, equipment, lot and step.
CoA from real data
Customer CoAs are generated from the actual measured values — not retyped from a binder.
Two-signature release
Where 211.165 / 111.123 require it, the kiosk enforces second e-sig before status moves to Released.
Curious how this lands in your environment?
Where AI actually earns its place.
AI does the sorting work of review-by-exception — reading every rolled-up step and putting the two that matter in front of the reviewer, with the CoA already drafted from real measured values.
- 01
Exception queue ranked, not just listed
AI ranks open exceptions by regulatory weight and product impact and groups repeats from the same line or lot, so the reviewer's first click lands on the thing worth ten minutes, not the thing worth thirty seconds.
Reviewers stop scrolling clean steps to find the two that aren't.
- 02
Deviation narrative drafted from step data
For any out-of-tolerance result routed to a deviation, AI writes the first draft of the narrative — equipment, lot, operator, timestamp, and any prior occurrence on the same line — for QA to edit and own.
Typical deviation write-up drops from ~40 minutes to a few minutes of review.
- 03
Customer CoA drafted from measured values
AI assembles the CoA from the batch's actual recorded results against the customer's specification template, so the reviewer checks a draft instead of retyping numbers from a binder.
AI never releases a batch and never applies the second signature that 211.165 or 111.123 require. It ranks, drafts and explains; the release decision is always made and signed by an authorised human.
One operator action — a complete, signed record.
What it leaves behind
- CAPA backlog visible per area
- Audit-trail entries on every status change
- Auditor portal — read-only login per batch
The clauses this satisfies
What buyers ask before they switch on QC release.
The rest of the platform this plugs into.
V5 isn't a bolt-on. Every module shares the same data, the same audit trail, the same operator. Pick where to look next.
Receiving
Scan-gated goods-in: matches PO, captures vendor CoA + lot, auto-quarantines, and prints a GS1-128 license plate before putaway. Built for 21 CFR 211 and FSMA 204.
Dispense
Live-scale dispense with tolerance bands and lot-genealogy capture at the gram. Wrong lot? The kiosk says no. Engineered for 21 CFR 211.101 and 111.40.
Production
MMR is locked and snapshotted into every WO. The eBMR fills itself in as operators work — temperature, hold time, in-process checks, all signed.
Got questions, or want to see it on your shop floor?
Ask V5 — our code-aware assistant — or spin up a workspace. Both are free.