Find what was already tested before you run it again.

Your LIMS (or your paper notes) tracks the sample.datin keeps each result with the specimen, machine and method behind it.

One result, traced back One result, a measured curve, with a coral thread from it back to what produced it: the specimen, the machine and the method. Result Specimen Machine Method

Running robots or automated benches?See robotic labs →

Can your agent tell these are the same sample?

Now imagine 100,000 of them.

Many questions, one sample Two lab tables name the same sample differently: a synthesis log with S-0042 and a table of GC results with s42, joined by a dashed line that asks: same sample? Around them lies a heap of more such pairs, from formulation logs, wear tests, instrument exports and spreadsheets, each with its own mismatched IDs, such as B-117 and b117, and its own question. Below the heap, the one sample, S-0042, as a node in a graph, linked to its synthesis step with Pd/C at 80 °C, its GC result of 91 %, its instrument, the GC, and its method. In motion, the first pair holds alone, more pairs pile in until the band is a heap of tables and questions, then the heap settles and the one sample lifts out below it. S-0042 Synthesis: Pd/C, 80 °C GC result: 91 % Instrument: GC Method Formulation log Base Batch PAO 4 C-301 PAO 6 C-302 results.xlsx sample wear c301 0.40 c302 0.44 same sample? Wear tests wear Specimen 0.42 mm N-61 0.37 mm N-62 Formulation log Batch Base n61 PAO 6 n62 PAO 8 same sample? Wear tests wear Specimen 0.29 mm W-2210 0.33 mm W-2211 results.xlsx ID COF wear 2210 0.09 wear 2211 0.12 same sample? Instrument export run sample_id r52 G-4 r53 G-5 Formulation log Batch Base G04 PAO 6 G05 PAO 6 same sample? results.xlsx COF ID 0.14 U8 0.09 U9 Wear tests Specimen wear U-08 0.37 mm U-09 0.29 mm same sample? Instrument export run sample_id r22 m_12 r40 m_13 Wear tests Specimen wear M-12 0.42 mm M-13 0.37 mm same sample? results.xlsx COF ID 0.13 Lot 19 0.11 Lot 20 Formulation log Batch Base L-019 PAO 8 L-020 Ester same sample? Wear tests Load Specimen 40 N P-9 40 N P-10 results.xlsx ID COF P09 0.13 P10 0.11 same sample? Formulation log Base Batch PAO 4 Y-402 PAO 6 Y-403 Instrument export sample_id run y402 r08 y403 r21 same sample? Formulation log Base Batch Ester E-55 PAO 8 E-56 Wear tests sample wear e55 0.38 mm e56 0.36 mm same sample? Instrument export run sample_id r40 D-12 r41 D-13 results.xlsx ID COF d 12 0.09 d 13 0.12 same sample? Instrument export run sample_id r21 M-3a r22 M-3b Wear tests sample wear M3 (a) 0.33 mm M3 (b) 0.35 mm same sample? Wear tests wear Specimen 0.37 mm A-220 0.29 mm A-221 Formulation log Batch Base a220 PAO 6 a221 PAO 8 same sample? Instrument export run sample_id r41 j14 r07 j15 Formulation log Batch Base J-14 PAO 8 J-15 Ester same sample? Formulation log Base Batch PAO 6 B-117 PAO 8 B-118 Wear tests sample wear b117 0.42 mm b118 0.37 mm same sample? Wear tests Load Specimen 20 N W-2031 40 N W-2032 results.xlsx ID COF wear 2031 0.11 wear 2032 0.14 same sample? Instrument export run sample_id r07 K12-01 r08 K12-02 results.xlsx Probe Ra K 12/1 0.21 K 12/2 0.19 same sample? results.xlsx Visc. Lot 46 Lot 5 68 Lot 6 Formulation log Batch Base L-005 Ester L-006 PAO 4 same sample? Synthesis log Catalyst T Sample ID Pd/C 80 °C S-0042 Pt/C 80 °C S-0043 GC results sample yield s42 91 % s43 87 % same sample?

56%

of design engineers' working time went into finding, receiving and passing on information.

Source

Sitting on old data?You've come to the right place.

datin makes hidden knowledge in your old data usable for AI.That knowledge informs today's decisions and gets you to market sooner.

Old results join today's record Diagram: old test results from past years join today's record. Each links to its specimen, machine and method. datin's agent answers from an old result and a recent one, with sources. Today Years ago Last year Specimen Machine Method Tested this before?

A result without its history is a guess.

Sample, machine and project, woven by every test Three threads run side by side over time: the sample, the machine and the project. Each changes on its own: the sample is cut, hardened and ground, a machine part is swapped, the project's method is revised and a new person joins. Every test is a thread woven over and under all three at the moment it ran. The result of one past test is traced back in coral to the state each thread was in when it ran. Sample Machine Project Test

The same structure holds across:

  • Materials
  • Chemicals
  • Lubricants
  • Coatings
  • Bearings
  • Gears and drivetrains
  • Automotive and e-mobility
  • Aerospace

Samples, machines and the lab itself change over time. datin traces all three side by side, and every test ties them together.

From the bench to the answer

  1. Record at the bench

    datin captures each step when it happens, with its specimen, machine and method.

  2. Trace any result

    Every step is its own record: when and where it ran, who or what ran it, what went in and out.

  3. Analyze on the record

    Run your analysis on the record itself, in the app.

  4. Add the next test without a database change

    Add a new test type as a new workflow, not a database project.

  5. Tables across tests

    See results from different tests in one table, with conditions as columns.

Your test pyramid

Bench tests cost little.Rig tests and approvals cost a lot.To run fewer expensive tests, start by linking the conditions behind every level.

Your test pyramid Test pyramid: many low-cost bench tests at the base, fewer rig tests in the middle, few costly system tests and approvals at the top, with their conditions linked across all three levels. Bench tests Rig tests System testsand approvals more tests higher costper test linkedconditions

Is datin right for your lab?

For one instrument, a general AI assistant reading a folder of files works fine.datin starts paying off as tests, machines and labs multiply.

datin is for you if...

datin is NOT for you if...

  • Processes

    Good fit: Steps change monthly

    Not a fit: Strict SOPs that never change

  • Materials

    Good fit: Variable, multi-method testing

    Not a fit: One standard test

  • Equipment

    Good fit: Regular calibration, changeable sub-parts

    Not a fit: Static instruments

  • Old data

    Good fit: Years of results you'd reuse if you could find them

    Not a fit: Nothing older than this year matters

Somewhere in between? A 15-minute call usually settles it.

Start where your lab is.

Have a LIMS?Keep it.

datin adds the history behind each result.

A cape for your LIMS The lab's existing LIMS, a plain box of small system tiles, wears a datin teal cape. It is tied at the box's two top corners and blows out to the right in the wind. LIMS

No LIMS yet?Skip straight to a better one.

Lab Workspace is your LIMS and ELN, without the IT project. Easy to use and flexible enough for any test: add your own specimen types, machines and fields.

  • Enter your way: draw a test as a mind map, fill in a template or write notes.
  • Read your way: on a map, in a table or from datin's agent, built right in. It cites its sources.
Add your own step in Lab Workspace An app window titled Lab Workspace, open on its Map tab next to Table and Agent, shows a chain of lab steps running top to bottom linked by teal lines from Specimen through Polishing, Machine and Test to Result, with the new Polishing step set in between Specimen and Machine and a soft teal ring round it. Lab Workspace Map Table Agent Specimen Machine Test Result Polishing

Thinking of building in-house?

Your IT team can.See where in-house builds usually break and why labs choose datin.

Questions

Does datin replace our LIMS or ELN?

No. datin sits beside them. No LIMS or ELN yet? datin's Lab Workspace fills the gap.

Can you bring in our old data?

Yes. In the first phase, our team brings in your lab history with you from existing exports. We start with the tests that hurt most. The first question takes the work. Each one after costs far less.

Where does datin run?

EU or US cloud, your servers, or your own cloud on request.

Is our data published?

No. It stays private and yours.

What happens to our data if we stop?

Your data and the models built on it stay yours to export. Your data stays readable without us.

Bring us the test that hurts most.

  1. 15-minute call
  2. Scoping
  3. One lab first, fixed fee

Deployed on-premises or in the cloud. Your data stays yours, and confidential.

Your contact

Dr. Nick Garabedian

Dr. Nick Garabedian

Founder & CEO, datin GmbH

Connect on LinkedInBook 15 min