{"id":902652,"date":"2026-05-02T20:59:05","date_gmt":"2026-05-02T20:59:05","guid":{"rendered":"https:\/\/example.com\/?p=1004"},"modified":"2026-06-10T02:34:02","modified_gmt":"2026-06-10T02:34:02","slug":"what-to-expect-getting-started","status":"publish","type":"post","link":"https:\/\/cri-one.com\/blog\/2026\/05\/02\/what-to-expect-getting-started\/","title":{"rendered":"What to Expect: Getting Started with Alchemy Probability Data"},"content":{"rendered":"

Three ways in<\/h2>\n

The system is designed so that the first useful query is reachable on day one, regardless of what your stack looks like. There are three primary access paths.<\/p>\n

1. The web interface<\/h3>\n

The fastest on-ramp. Start the local API server and open the bundled web page in a browser.<\/p>\n

python bluray_data_api.py\n# or use the Windows shortcut\nSTART_WEB_INTERFACE.bat<\/code><\/pre>\n
From there, the interface exposes tabs for compound search, transformation lookup, and recipe generation. No coding needed for the first pass.<\/p>\n
2. The Python API<\/h3>\n
For teams already in a notebook or pipeline, the Python reader is the natural fit:<\/p>\n
from bluray_data_reader import BluRayDataReader\n\nreader  = BluRayDataReader(\"bluray_data\")\nresults = reader.search_compounds(['H', 'O'], max_results=10)\nrecipe  = reader.generate_invention_recipe(['H', 'O'])<\/code><\/pre>\n
Same data, fully programmatic. This is the path most R&D teams will end up on once they move past initial exploration.<\/p>\n
3. The interactive CLI<\/h3>\n
For terminal-driven workflows or quick batch experiments:<\/p>\n
USE_DATA.bat\n# or\npython bluray_data_reader.py interactive<\/code><\/pre>\n
The CLI is also the foundation for batch processing \u2014 pointing the reader at a list of targets and letting it work through them unattended.<\/p>\n
What deployment looks like<\/h2>\n
Realistic timelines, depending on how deep you need to go:<\/p>\n
    \n
  • Basic deployment (1\u20132 days)<\/strong> \u2014 web interface or Python API, single machine, exploratory use.<\/li>\n
  • Production deployment (1\u20132 weeks)<\/strong> \u2014 server setup, configuration, internal access controls, validation runs.<\/li>\n
  • High-performance deployment (2\u20134 weeks)<\/strong> \u2014 dedicated hardware, optional accelerator integration, tuned storage for the full 2,609-file corpus.<\/li>\n
  • Custom integration (1\u20133 months)<\/strong> \u2014 embedding the system into an existing research platform or data pipeline.<\/li>\n<\/ul>\n
    What the package contains<\/h2>\n
    When you receive the system, it arrives as a self-contained handoff:<\/p>\n
      \n
    • Complete data package (handoffs\/bluray_data\/<\/code> \u2014 2,609 files).<\/li>\n
    • Processed data outputs (handoffs\/alchemy_data_v2_output\/<\/code>).<\/li>\n
    • Web API server, Python reader library, and web interface.<\/li>\n
    • Core system components and data-access modules.<\/li>\n
    • 61 documentation files covering quick starts, architecture, generation, and validation.<\/li>\n<\/ul>\n
      What to expect, plainly<\/h2>\n
      You should expect a working query within an hour of unpacking, a useful exploratory session within a day, and a real production footprint within a couple of weeks. The system is positioned as a complete handoff \u2014 not a research prototype \u2014 and is priced accordingly at $10B<\/strong> for the full data and tooling package.<\/p>\n
      If you have followed the series this far, you have the premise (post 1), the data (post 2), the use cases (post 3), and now the path to actually using it. From here, the work is yours.<\/p>\n
      <\/p>\n
      \n