WHERE
DATA
MEETS 
NEWS & UPDATES
HXMX will present at the WTGS 2025 Fall Symposium!
September 16-18, 2025
Midland, TX
Our talk, “From the DJ to the Williston: Advancing Formation Tops Fingerprinting with Sights on the Permian,” highlights how we’re combining signal-based pattern recognition with stratigraphic logic to deliver faster, more consistent, and transparent tops picking results. It has been a joy collaborating with CoreGeo on this paper alongside James Hawkins J. Alexandra Sloan, MS, MBA Clayton Mack.
HXMX Unlocked the Second Tranche of their RRB Grant
June 20, 2025
Tulsa, OK
HXMX unlocked the second tranche of their Rose Rock Bridge grant by establishing a Tulsa office and hiring Tulsa staff.
Utility Patent Application Filed: Automated Identification of Serial or Sequential Data Patterns by Marker Fingerprinting
January 15, 2025
Marker Fingerprinting is a proprietary method developed by HXMX for identifying patterns in well log and geological data. By converting key features into compact, searchable fingerprints, the system rapidly detects matching patterns across datasets, flags inconsistencies, and provides confidence metrics for interpretation. This reduces manual effort and increases the precision of subsurface correlations.
HXMX Receives the First Tranche of their RRB Grant
October 11, 2024
Tulsa, OK
First funding for the Rose Rock Bridge grant was received October 11, 2024 and provides invaluable cash support as well as legal and design assistance and access to the Microsoft for Startups program.
HXMX Accepted into the 2024 RRB Cohort
August 12, 2024
Tulsa, OK
Rose Rock Bridge graciously accepted HXMX into their 2024 cohort, providing access to the VentureWell startup eduction program and the opportunity to gain financial and professional support.
Provisional Patent Application Filed: Automated Identification of Serial or Sequential Data Patterns by Marker Fingerprinting
January 15, 2024
HXMX filed their first provisional patent application on January 15, 2024 to protect a range of pattern recognition techniques.
PAPERS & PUBLICATIONS
Advancing Formation Tops Fingerprinting: Expanding Applications and Insights from the DJ to the Williston Basin
June 9, 2025
HXMX and Core Geologic worked jointly on this paper which extends the Tops Fingerprinting approach from the DJ to the Williston Basin. Algorithmic picks were compared to those of five geoscientists across 23 formation tops and 9,000 vertical feet of section. In 17 cases (74%), the algorithm matched or outperformed human interpreters, demonstrating consistent performance across most intervals.
Tops fingerprinting: Revolutionizing well log analysis with music recognition technology
August 27, 2024
You hear a song but can't quite place it. You pull out your phone and use a music recognition app to identify the song immediately. That process is hashing a short snippet of signal and identifying where it comes from in a database of hashes from millions of songs in seconds. Can that approach be adapted to help with picking and checking tops on well logs? The short answer is yes!
Adapting Music Recognition Technology for Tops Picking and Quality Control
June 17, 2024
HXMX presented DJ work done with the help of the Innov8x program at the Colorado School of Mines and experts from Ovintiv, LLOG, Meagher Energy Advisors, and Oxy on the Tops Fingerprinting technique at URTeC in 2024. This paper dives into the heart of the technique and the implications for both tops and fault picking. It also describes the technique first used to determine consistency of human picks and compares that variance to that of the algorithm. The proof-of-concept test conducted in the DJ Basin involved one reference well and nine target wells, which were picked both by six geologists and by the algorithm.
An Industrial Strength Search Algorithm
January 1, 2003
This article provides a fascinating overview of the elegant pattern matching technique used by Shazam. Written by Avery Wang, now at Apple, this approach inspired the tops picking approach used by HXMX. This method is very fast and does not require training in the AI sense.
%20--%3e%3cdefs%3e%3cstyle%3e%20.st0%20{%20fill:%20none;%20stroke:%20%23fff;%20}%20%3c/style%3e%3c/defs%3e%3cpath%20class='st0'%20d='M646.6,704.9l-40.5-36.5-287-72.7,4.9-180.4,139.7-97.7,199.7,138.9-71,19.3,54,229Z'/%3e%3cpath%20class='st0'%20d='M319.2,595.7l144.7-278,199.7,138.9-57.4,211.8,40.5,36.5-322.5-289.5,268.5,60.5-273.4,119.8Z'/%3e%3c/svg%3e)