PROFILE

Holy Grail

What we do:

Super intelligence for research and optimization problems

Why:

Deep Learning methods can accelerate R&D and process optimization massively but they are too complex to implement and use

Features

  • General probabilistic framework - designed to perform well with small and noisy data
  • Traditional machine learning - find correlations between features and targets
  • Multi-variable and multi-target - allows simultaneous optimisation of multiple targets
  • Active learning - predict features that might generate the targets you are looking for (used for Design Of Experiment)
  • Unbiased - there is no guesswork or plot analysis, the models are built on empirical data and not influenced by human biases
  • Easy to use - no coding skills

CONNECT

David Pervan
david@holygrail.ai
1 415 767 6958

PROFILE

Holy Grail

INDUSTRY

R&D, Optimization

COMPETITIVE ADVANTAGES

  • All models are probabilistic in nature so every output comes with its associated uncertainty
  • Active Learning
  • Works with small datasets (below 1,000 data points)
  • Ease of use - no data science or coding experience required

TARGET MARKETS

  • R&D departments
  • Manufacturing
  • Process optimization
  • Universities

USE CASES

Equipment calibration

  • A research laboratory is able to calibrate their equipment (e.g. 3D Printers) acquiring only 20% of the entire variable space they used to use to calibrate.
  • This reduced their calibration time by ~40%

Design of Experiment (DOE)

  • An R&D department was trying to develop a new catalyst but was unable to progress because they were unable to make sense of their noisy experimental data
  • Using their physical experimental data and density-functional theory (DFT) results, our software was able to guide the experimental work towards the most likely material

REFERENCES

  • Client industries include: energy storage, manufacturing, bioplastics, plant-based meats, cell-grown meats, biofermentation processes, bioreactor design

COMPANY

  • Founded: 2019
  • Raised pre-seed investment from Deep Science Ventures, London, UK
  • Raised investment from the Silicon Valley-based Y Combinator and successfully went through the Y Combinator Summer 19 batch
  • Headquaters: San Francisco, California

CONNECT

David Pervan
david@holygrail.ai
1 415 767 6958