Application Refactoring · Customer Service Automation · IT Operations Automation
cult.fit wanted teams to solve operational and product challenges without adding process complexity. Fragmented developer tools hindered shared engineering practices, data requests bottlenecked at a central analyst team, and the company sought more personalized member experiences and lower-cost customer support.
Reported impact
cult.fit reports more than 20 internal applications built by non-engineering teams in recent months, over $100,000 in annual software savings from rebuilding an internal capability, and a 50%+ reduction in customer support costs. It also reports shipping a social feed in two months instead of the more-than-a-year estimate for the previous approach.
cult.fit standardized software development around Codex, with shared repository structures, skills, testing, code review and feedback loops; its SRE agent aggregates observability-platform and log information, investigates incidents and can prepare a pull request for engineer review. Security agents address vulnerabilities, while on-call agents investigate recurring production issues. Beyond engineering, teams use Codex with company data to analyze information and build internal applications and workflows, including HR performance-review coordination. The OpenAI API and GPT Realtime support the AI Trainer's two-way voice coaching and workout logging. For customer support, AI agents are built into Sprinklr and connected to backend systems to assess penalty-waiver requests using local weather and class-attendance signals. The page does not identify underlying storage or deployment topology.
For workouts, members interact by voice with the AI Trainer, which uses GPT Realtime through the OpenAI API to provide coaching cues and support workout logging. For support, Sprinklr-hosted agents check local weather and class-attendance patterns against cult.fit backend systems to assess a member's penalty-waiver request. Codex also works with company data for team analysis and app building.
Cult.fit built these agents into its existing Sprinklr support platform and connected them to its backend systems.
Cult.fit now runs the capability on its own application, avoiding more than $100,000 in annual software costs.
More than 20 internal applications have been built by non-engineering teams in recent months.
The latest version is designed around a voice-first workout experience. Using GPT Realtime through an OpenAI API integration, the trainer can provide coaching cues and support workout logging through two-way voice interaction.
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
KPMG Sample Nov 6, 2026 Fortune 50 healthcare organization A complex post-merger environment left the healthcare organization with an over-customized ServiceNow platform, technical debt, inconsistent processes, and multiple service systems that made support difficult to… 10,000 knowledge articles and 500 catalog items ServiceNow · ServiceNow AI Control Tower · ServiceNow Virtual Agent 
OpenAI Sample Oct 9, 2026 Asana Asana needed to cut the cost and latency of its StackAI browser agent so it could offer customers more capable models for browser-driven workflows without unsustainable operating expense. $0.47 average estimated model cost with the optimized GPT-6.1 Sol workflow Codex · Command · GPT-6 Astra OpenAI Sample Oct 8, 2026 Braintrust Braintrust wanted to respond to customer feature requests faster and run more engineering experiments, but slower coding tools and backlog-driven workflows made it costly to turn feedback into working prototypes in real… 50% of the Braintrust team moved to Codex in one month Codex · GPT-5.5 