---
title: "AI agents for R&D, every claim sourced before your review"
description: "Duale AI agents read papers, patents, and case files to produce sourced evidence maps, compound dossiers, and case pre-reads for R&D review."
lang: en
lastUpdated: 2026-10-23
url: https://duale.ai/en/use-cases/research-and-development
---

<Hero
  eyebrow="R&D"
  title={
<>
  AI agents for R&D, <AccentText>every claim sourced before your review.</AccentText>
</>
}
  subtitle="Hand Duale AI a research question, a candidate or a case file. You get back one result in the format you set. One Task reads the papers, patents and files in focused branches and runs in the background."
  primaryCta={{ label: "Plan your first use case with us", dialog: "contact" }}
  secondaryCta={{ label: "See pricing", href: "https://duale.ai/en/product/pricing.md" }}
/>

<Section title="Literature synthesis, what the evidence supports and what it leaves open" subtitle="Your team sees how strongly the sources back each claim, with the quoted passage beside it.">
  **You hand over:** "Map the evidence on recycled content in packaging film for Monday's project review. How does
  strength change, what offsets the loss, and below which share does it hold?"

  **How the Task splits it:** one branch per sub-question searches and reads the published papers with the Web research
  built into [Tools](https://duale.ai/en/product/tools.md). A review branch then checks every claim against the passage it quotes. You can
  watch the same fact-check in the [3-minute live demo on the use cases page](https://duale.ai/en/use-cases.md).

  **You get back:** one evidence map. Each claim comes with its quoted passage, its source and date, and how well the
  sources support it. The gaps that no source answers come last.

  **Illustrative example: an evidence map on recycled content in packaging film, ready for Monday's project review. The
  papers are fictional.**

  | Claim                                         | Quoted passage                                         | Source                      | Status                  |
  | --------------------------------------------- | ------------------------------------------------------ | --------------------------- | ----------------------- |
  | Tear strength falls as recycled content rises | "Tear resistance fell by 18% at 50% recycled content." | Paper A (2025), section 3.2 | Supported by two papers |
  | An additive offsets most of that loss         | "With 5% additive, the loss fell to 4%."               | Paper B (2024), table 4     | One study; to confirm   |
  | Below 20% recycled content, strength holds    | No passage found                                       | None                        | Gap: no study tests it  |

  **Your team decides:** the research direction, and which gap to test first.
</Section>

<Section title="Candidate compound dossiers, checkable line by line" subtitle="Each candidate reaches your portfolio review with a dossier that separates calculated values from sourced ones.">
  **You hand over:** "Prepare the dossier on this candidate additive for marine coatings: its properties, the prior art
  and our earlier tests, for the portfolio review."

  **How the Task splits it:** separate branches read the published literature, the patents and your earlier test
  reports, kept in a [Library](https://duale.ai/en/product/libraries.md). Code execution calculates the properties you ask for. The dossier
  marks which values were calculated and which come from a source.

  **You get back:** one dossier with the calculated properties, the prior art with each source dated, your earlier
  results that bear on the candidate, and the open questions.

  **Illustrative example: a dossier on a candidate additive for marine coatings. The compound and the patents are
  fictional.**

  | Section               | What comes back                                                             | Source                                                         |
  | --------------------- | --------------------------------------------------------------------------- | -------------------------------------------------------------- |
  | Calculated properties | Molecular weight of 312 g/mol; low estimated water solubility               | Calculated with code execution from the structure you provided |
  | Prior art             | Two patents claim close structures for marine coatings; one expired in 2024 | Patent A and Patent B, each with its date                      |
  | Your earlier tests    | A close analogue failed your adhesion test in 2023                          | Your March 2023 test report, page 6                            |
  | Open question         | Does the candidate pass the adhesion test that its analogue failed?         | To test in your lab                                            |

  **Your team decides:** whether to advance the candidate, test it further, or drop it.
</Section>

<Section title="Case pre-reads, each deviation sourced and set against your guidelines" subtitle="Your review board opens each case with the questions to settle already listed.">
  **You hand over:** "Prepare the pre-read on the returned heat gun case for tomorrow's product safety board."

  **How the Task splits it:** one branch per section of the case file reads it against the guideline sections that apply.
  The Task gathers what they find into one pre-read.

  **You get back:** a summary of the case, the quoted guideline passages that apply, each deviation with its source in the
  file, and the questions for the board.

  **Illustrative example: a returned heat gun, a fictional product, before tomorrow's product safety board.**

  | Case file section | Guideline passage                                                                      | Finding                                                             | For the board                           |
  | ----------------- | -------------------------------------------------------------------------------------- | ------------------------------------------------------------------- | --------------------------------------- |
  | Incident reports  | "Three reports of the same failure within 30 days open a safety review." (section 4.1) | Four overheating reports within 22 days                             | Review opened as required               |
  | Test results      | "Surface temperature stays below 70 °C after 10 minutes of use." (section 6.3)         | One lab test recorded 78 °C after 10 minutes                        | Retest the current batch?               |
  | Design changes    | "A new supplier for a heating part requires a new test." (section 7.2)                 | The heating element supplier changed in May; no new test is on file | Deviation: ask engineering for the test |

  **Your board decides:** the outcome of the case and the actions that follow.
</Section>

<Section title="The capabilities behind each research result" subtitle="Each one is set up once on the Platform and serves your other use cases too.">
  <CardGrid columns={2}>
    <Card title="Libraries" icon="reader" href="https://duale.ai/en/product/libraries.md">
      Keep your test reports and guidelines in Libraries. Each result cites the document and the page it used.
    </Card>

    <Card title="Tools" icon="code" href="https://duale.ai/en/product/tools.md">
      Web research reads published papers and patent records, each with its date, and code execution calculates the
      values you ask for.
    </Card>

    <Card title="Model routing" icon="mixer-horizontal" href="https://duale.ai/en/product/model-routing.md">
      Each branch runs on a model you enabled that suits its work: long papers on one, calculations and synthesis on
      another.
    </Card>

    <Card title="Deployment" icon="globe" href="https://duale.ai/en/product/deployment.md">
      Your unpublished research and your case files can stay in your own infrastructure.
    </Card>
  </CardGrid>

  [See all use cases](https://duale.ai/en/use-cases.md).
</Section>

<Section title="Before your first synthesis" subtitle="The facts your security, finance and procurement colleagues check.">
  <CardGrid columns={3}>
    <Card title="One fixed price per Task" icon="check">
      A standard Task costs €0.10 and a fast Task €0.20, excluding VAT, whatever the number of branches or the duration.
      [See pricing](https://duale.ai/en/product/pricing.md)
    </Card>

    <Card title="Your providers, your contracts" icon="layers">
      Your model providers bill you directly for inference.
    </Card>

    <Card title="Your choice of hosting" icon="globe">
      Use the managed service, hosted in the European Union, or run the Platform in your own infrastructure.
    </Card>

    <Card title="Encrypted storage" icon="lock-closed">
      All stored data is encrypted, from Library files to database backups.
    </Card>

    <Card title="Unlimited agents" icon="person">
      Add agents as your teams and use cases grow.
    </Card>

    <Card title="Your content remains yours" icon="file-text">
      Duale AI does not reuse your content to train a model. [See subprocessors](https://duale.ai/en/legal/subprocessors.md)
    </Card>
  </CardGrid>
</Section>

## R\&D questions

### Can a synthesis weigh our unpublished findings against the literature

Yes. The evidence map sets each finding from your internal reports beside the published claims it supports or
contradicts. [See how Libraries keep your reports](https://duale.ai/en/product/libraries.md), then [see how Web research in Tools
brings in public sources](https://duale.ai/en/product/tools.md).

### Can the dossier use our own calculation code

Yes. The dossier's calculated values come from the code your team already trusts. [See how your own code becomes a
Tool](https://duale.ai/en/product/tools.md).

### Can a pre-read apply our internal rules as well as published standards

Yes. The pre-read applies both, and the board sees whether each passage is an internal rule or a published standard.
[See how Libraries keep your guidelines](https://duale.ai/en/product/libraries.md).

### Can one project's synthesis serve the next one

Yes. Keep each finished evidence map in a Library, and the next synthesis starts from it, cites it and adds what was
published since. [See how Libraries keep your past work](https://duale.ai/en/product/libraries.md).

<Cta title="Choose the first synthesis, dossier or pre-read to hand over." primaryCta={{ label: "Plan your first use case with us", dialog: "contact" }} secondaryCta={{ label: "See the due diligence Task", href: "https://duale.ai/en/use-cases/finance/due-diligence.md" }} />

## Related content

- [AI due diligence from press and data room to one memo](https://duale.ai/en/use-cases/finance/due-diligence.md)
- [Legal teams: every point quoted, ready for decision](https://duale.ai/en/use-cases/legal.md)
- [Finance: every memo and report sourced before its meeting](https://duale.ai/en/use-cases/finance.md)
- [Sales: bids and briefs sourced before the deadline](https://duale.ai/en/use-cases/sales.md)
- [Customer service: calls reviewed, trends traced to accounts](https://duale.ai/en/use-cases/customer-service.md)
- [IT and security: alerts classified with evidence](https://duale.ai/en/use-cases/it-and-security.md)

---

## Sitemap

See the full [Markdown sitemap](https://duale.ai/sitemap.md) for all pages.