• For buyers
  • Docs
Sign inSign up
Data opportunities

Science and R&D

Semiconductor & materials company

License the run records that tie each recipe to a measured result.

We're looking for companies that make semiconductors, devices and advanced materials, and that keep careful records of what they built and how it performed. The most valuable data links each build's recipe or process settings to its measured result, with failed builds and repeat runs kept. Lab notebooks that explain why a build failed are especially useful. Labs and startups that have wound down qualify.

Paid to the company
Up to $1,000,000Depends on run count and how well runs link to results
Referral bonus
Up to $100,000
Company size
10–2,000 employees
Location
Anywhere
History
Several years of runs
Typical industries
Fabs and foundries, device makers (solar, photonics, quantum, power electronics), fabless chip designers, process and test equipment makers, materials and battery R&D labs
Start selling

Systems the data usually lives in

  • MES
  • SPC / yield systems
  • LIMS
  • Electronic lab notebooks
  • Cadence
  • Synopsys
  • OSIsoft PI

Requirements

  • Run or build records the company owns, with recipe or process settings linked to measured performance
  • 1,000+ runs, builds, wafers or experiments
  • Every build kept, failed ones included, with dates
  • Data that hasn't been published

Nice to have

  • Lab notebooks with notes on why builds failed
  • The same recipe built again on different days
  • Measurements under more than one condition
  • Yield, metrology and equipment logs
  • Chip or device designs with their verification and test results

What is included

  • Run and build records with measurements
  • Lab notebooks
  • Process, yield and equipment data
  • Design and test files, where available

Not a fit

  • Data owned by a customer or foundry partner
  • Published papers and datasheets only
  • Licensed third-party IP

Scoping questions

  • What run or lab records do you keep, and in which systems?
  • Roughly how many runs or builds, over how many years?
  • Do you keep failed builds and repeat runs?
  • Do you own the data, and has any of it been published?

Who is best placed to refer

Process and device engineers, fab and lab managers, materials scientists, and founders of wound-down hardware startups.

Questions about data opportunities

  • Company size: 10–2,000 employees
  • Location: Anywhere
  • History: Several years of runs
  • Typical industries: Fabs and foundries, device makers (solar, photonics, quantum, power electronics), fabless chip designers, process and test equipment makers, materials and battery R&D labs

Companies that qualify have:

  • Run or build records the company owns, with recipe or process settings linked to measured performance
  • 1,000+ runs, builds, wafers or experiments
  • Every build kept, failed ones included, with dates
  • Data that hasn't been published

Not a fit:

  • Data owned by a customer or foundry partner
  • Published papers and datasheets only
  • Licensed third-party IP

Up to $1,000,000. Depends on run count and how well runs link to results. Whoever refers the company can earn a referral bonus of up to $100,000.

Create a vendor account and list your data. Then apply for any open buyer opportunity it fits. The DataVendor team reviews the data you have, helps you prepare it for sale, lists it on the marketplace, and helps you sell it to the AI companies that want it.

To move faster, have answers to the questions the team asks first:

  • What run or lab records do you keep, and in which systems?
  • Roughly how many runs or builds, over how many years?
  • Do you keep failed builds and repeat runs?
  • Do you own the data, and has any of it been published?

A kind of company data that AI companies on DataVendor want to license. Each opportunity says which companies qualify, where the data usually lives, and the most a company can be paid. That amount is an upper bound, not a quoted price.

The DataVendor team reviews the data you have, helps you prepare it for sale, lists it on the marketplace, and helps you sell it to the AI companies that want it. You decide what you license and to whom.

No. The DataVendor team helps you prepare the data for sale, including removing personally identifiable information (PII) and secrets, with the scope agreed for your data. Tell us about any sensitive fields or files when you apply.

Yes. The systems on an opportunity are where that data usually lives, not a requirement. The DataVendor team reviews what you have and helps prepare a package buyers can use, even when a system has no direct connector.

Nothing up front. Listing, review, and preparation help cost nothing; DataVendor takes only a transaction fee when you withdraw your earnings. See the current fees.

Sign up, then share your referral link from Refer a vendor in your account menu with a company that fits. If the company signs up through it and a buyer deal involving its data closes within 180 days, DataVendor pays you a referral fee. The fee comes out of DataVendor's share, never the company's payout. The referral bonus on an opportunity is the most that referral can pay. Terms are in the DataVendor referral agreement.

You retain ownership by default and license the agreed uses. The agreement sets the rights and any exclusivity. Buyer offers identify the organization so you can decide whether to accept; buy-now purchases remain anonymous. If buyer approval or use restrictions matter to you, discuss them with us before publishing. Read the seller agreements before accepting them.

Estimator tools

  • Codebases
  • Company data

Resources

  • For buyers
  • Docs
  • Data opportunities
  • Blog
  • Contact
  • Careers

Legal

  • Privacy
  • Terms
  • Trust

Social

  • GitHub
  • X
Provided by
  • For buyers
  • Docs
Sign inSign up