Energy / Energy data services
TGS: Searching knowledge, guiding troubleshooting and creating support tickets for seismic-processing failures
- Company
- TGS
- Country
- Multiple countries
- Adoption stage
- In operation
- Source published
- Date basis
- The date the source was published. It can differ from the date adoption started.
- How the source was checked
- Read the full source text
The work problem
TGS runs seismic-imaging jobs across time zones, but expert support followed local business hours. Knowledge was scattered across Jira, Confluence and SharePoint, leaving high-performance computing idle while staff searched for answers.
Technology and data
TGS operates Scout on AWS Quick. It enforces source permissions during retrieval and cites supporting documents. With user confirmation it creates or updates Jira tickets; high-impact actions require human review. It withholds definitive answers when evidence is insufficient.
Results
TGS and AWS reported that, across 35,000 responses from March through October 2026, 99% were citation-grounded and users rated 70% accurate. Time to discover a solution fell from two to three hours to under three minutes. This measures discovery, not total recovery time.
Limits and open questions
Pre-launch baselines were not sufficiently captured by error class. Citations do not establish correctness or successful automatic recovery.
Sources
- Always-on, grounded, and going agentic: How TGS built Scout on AWS QuickAWS / TGS, Accessed
Compiled from public sources. These are not results from ATF Works customers.