Instrument printouts into your lab record, with the analyst in charge.
Captured at the bench, read into the record, sent on. A person approves what ships.
Capturethe printout at the bench, by itself
Readevery field, linked to its spot on the slip
Reviewonly what is flagged, then approve
Sendapproved values on to the record, the warehouse, the library
Keepthe slip image with the result, for good
What a bench produces today
The result is printed. It is not recorded.
A bench instrument measures the sample and prints the result on a strip of thermal paper. That is where the data stops: nothing can be released until the values are in the lab record - the LIMS, the ELN or the quality system. And the paper itself fades, while a GxP record has to stay readable for years.
Reading an instrument printout into a record is not a new idea. Instrument interfaces, scan-to-LIMS tools and middleware have done pieces of it for years. AI Intime is one way to do it - built for labs that need it to run inside their own boundary, to read any printout from any instrument, and to keep a person in charge of what is recorded.
Captured by itself
Captured without a keyboard.
Capture is the easy half. What matters is what happens between the capture and the record.
Bench scanner or camera
A fixed camera or a sheet-fed scanner at the bench. The slip is captured the moment it is torn off; nobody opens a file.
Watched folder
Point AI Intime at the folder your scanners already drop into. New files are picked up as they land.
Email in
Slips photographed on a phone and sent to a lab mailbox arrive in the hub by themselves.
SharePoint library
If your scans already live in a SharePoint library, the connector indexes it and keeps it in sync.
Instrument export, where it exists
If the instrument can export over USB, serial or the network, we take the file. If it can only print, we read the print.
See it work
From thermal slip to lab record, in a minute and a half.
Capture · read · review · approve · post. Demo data; no customer or instrument shown.
Sent on
Sent where it is needed.
Approved values do not sit in AI Intime. They go out, through the same connectors the platform uses everywhere else.
The lab record
LIMS, ELN or your quality system - the system your release process reads today. AI Intime writes to it through an interface adapter; nothing is replaced.
Data warehouse
Every approved result streams to your warehouse or lakehouse for trending, control charts and audits, with the slip image reference beside it.
SharePoint
The slip image and its approved values are filed to the library your QA already audits from, as the true copy.
Teams or Outlook
An exception that needs a decision, or a batch whose results are all in, posts to the channel or inbox that owns it.
Anything else
A REST endpoint and an MCP tool layer for the systems we have not named.
On the platform todaySharePoint · Outlook · Teams · database change-data capture (PostgreSQL, MySQL, SQL Server, MongoDB) · REST upload API · MCP tool layer
The approach
Three things a generic tool cannot give you.
Tap a card to read it. One at a time.
Reading the slipField by field, linked to the slip, with a person deciding the doubtful ones
A generic photo-to-text tool reads the whole slip in one gulp and hands back a block of characters: no fields, no link to where each value sat, and a faded digit guessed as confidently as a clear one
AI Intime recognises the instrument's layout and reads each field on its own - sample, parameter, result, units, time, instrument, operator - into the fields your record already has
Every value stays linked to its spot on the slip image, so the analyst checks against the source, not against memory
Faint or doubtful digits are flagged for a person to decide, beside the slip. The analyst's review is the documented second check
Capture and hand-off are automatic; the review is the only step that waits for a person
OutcomeOne human step, and it is the record. The judgment stays with the analyst, and the record shows who decided what.
DeploymentOn-prem. Air-gapped, if that is your standard.
Generic AI tools are cloud services - a non-starter for regulated lab data and batch records
AI Intime runs inside your infrastructure: your site, your private cloud, or fully air-gapped - and the connectors to your LIMS, warehouse and SharePoint run there too
We support containerized deployment by your own team, to the standard of our most restricted deployments
OutcomeZero data leakage. Results, slips and the record never leave your boundary.
Sovereign AIYour AI, on your terms - no lock-in to a model, a vendor or a price
Choose the AI model, and change your mind later. Providers retire models on their own schedule; here the model is a part you swap, and your slip capture keeps working exactly as it did
Keep control of cost. Cloud token prices change without asking you; run on a right-sized or self-hosted model and your cost per slip is a number you set, not one you receive
Keep control of where it runs. Your site, your cloud, or air-gapped - and if a regulator or a jurisdiction rules a provider out, you move and lose nothing
Keep your knowledge yours. Your instruments' layouts, your field names, your corrections stay with you - never inside a vendor's model
OutcomeFreedom of action. You decide what changes, when, and on whose terms.
What it unlocks
Release at the speed of review, not typing.
Captured by itself
Scanner, folder, mailbox or library - the slip reaches the hub without anyone opening a file. Capture is the easy half; what matters is what happens between the capture and the record.
The slip becomes a record
Sample, parameter, result, units, time, instrument and operator - read from the printout into fields your record already has, each one linked to its place on the slip image.
Faint print, flagged
Thermal print fades and a 3 can look like an 8. A doubtful digit is marked for a person to decide, beside the slip, rather than guessed.
Every change, logged
Who captured the slip, who changed which value, who approved. The trail is kept with the record, not in someone's memory.
A lasting copy
Thermal paper fades; the record has to last. The slip image is filed with the result as a true copy, so the paper no longer has to be photocopied to survive.
Every instrument, every slip
Different instruments print different slips. Each layout is learned once, reviewed by a person, and held for every future printout from that model.
Sent by itself
Approved values go to the LIMS, the warehouse and the SharePoint library through the same connectors the platform uses everywhere else - not into another database beside them.
On-prem, or air-gapped
Lab data is regulated and confidential. AI Intime runs inside your own infrastructure - on your site, or fully air-gapped - with the connectors running there too.
No model lock-in
The AI model is a part you can swap. Change providers, go self-hosted, or move when a regulator says so - your capture keeps working exactly as before.
The review flow
A person stays in charge of what is recorded.
Extraction is automatic; judgment is not - "Human in Loop" is our design philosophy, and in a GxP lab it is also the second check.
1AI
The slip is captured at the bench - from the scanner, the folder, the mailbox or the library.
2AI → You
The instrument layout is recognised and every field on the slip is read into the record.
3AI → You
Anything faint or doubtful is flagged beside its spot on the slip.
4You
Check the flagged values against the slip; correct, then approve. That is the second check.
5AI
Approved values post to the LIMS and stream to the warehouse; the slip image is filed to SharePoint as the true copy.
Other tools do pieces of this. Three things set ours apart.
Any instrument, any printout
Direct interfaces cover the instruments that have them. We read the paper from the ones that do not, and the same review screen covers both.
Inside your boundary
On-prem or air-gapped, with the connectors to your LIMS, warehouse and SharePoint running there too. Nothing goes to a cloud OCR service.
One human step, and it is the record
The analyst's review beside the slip is the second check your procedure already asks for, with who-changed-what kept with the result.
Who it's for
Built for the people who turn a printout into a released batch.
Teams
QC lab managers and analysts
QA and data integrity
Lab IT and the owners of the lab record
Data engineering, as the owners of the connectors
Pharma and biotech first; water and chemicals use the same bench instruments
Where it runs
Entirely inside your infrastructure - on your site, or fully air-gapped. The slips are your batch records - they never leave your boundary, and neither do the LIMS, the warehouse or the SharePoint library they feed. And it is LLM-agnostic: the models sit behind the platform, replaceable without touching the capture or the connectors.
See it on your own printouts.
Bring a handful of slips - as they came off the instrument, faded ones included - and watch the values land in your record with the analyst's review in the loop. Tell us where the values need to go, and we will show that too. No form, no gate.