SAN JOSE — BlueSprint Labs, a pre-seed accelerator billing itself as “the world’s first evidence-based pathway to grappling excellence,” launched its inaugural Blue Belt Sprint Cohort this week, promising investors and applicants that a structured 12-week program could “compress the traditional blue belt timeline from years of arbitrary mat time to a fully shippable outcome.”
By Thursday morning, five of the fourteen cohort members had been choked unconscious. Only two of them tapped.
“We ran the numbers,” said Brayden Kettler, 33, founder and CEO of BlueSprint Labs and a self-described “blue belt-adjacent practitioner” who has been training for six months. “The average time to blue belt is two to four years. That’s a resource allocation problem. We built a system.”
The system, as outlined in the program’s 47-slide pitch deck — available to accredited investors upon NDA execution — was a proprietary framework Kettler calls “Agile Grappling™”: a combination of two-week sprints, daily standups, and “outcome-based drilling” designed to “ship a technically competent practitioner” within the program window.
Week One, according to the cohort’s retrospective Notion document — titled Sprint 1 Retro: What Went Well / What Didn’t / Action Items — did not go well.
“We had a great standup,” said Dax Reinholdt, 29, a former product manager from a now-defunct Series B logistics startup and a member of the cohort. “Everyone checked in, flagged blockers, said what they were working on. Then we had to actually roll.” He paused. “Dax Reinholdt is blocked on armbar penetration and waiting on a dependency that won’t stop moving.”
Reinholdt was referring to his training partner, Tomás Olvera, a 44-year-old brown belt and volunteer instructor at the host facility, Cali Ground Warfare in Cupertino.
Olvera, who has been teaching jiu-jitsu for eleven years, said he was recruited through a professional networking message from Kettler describing the opportunity as a “high-growth coaching engagement in an emerging grappling-outcome space.”
“I thought it was a normal class,” Olvera said, wearing a gi that had been labeled, by Kettler, with a laminated badge reading Human Training API v2.1. “Then they started asking me to ‘accept their pull requests.’ I told them that is not how jiu-jitsu works. Then Dax tried to explain the pull request metaphor to me, and I put him in a triangle.”
Kettler acknowledged in a Slack message to the cohort — reviewed by this reporter — that “partners cannot be iterated on in the traditional software sense,” but insisted this was a “known risk” that had been “flagged in the risk register and assigned a mitigation workstream.”
The mitigation workstream was a whiteboard exercise.
By Week Four, the cohort had completed two sprints, collected 1,400 rows of drilling data in an Airtable base Kettler called the “Technical Debt Dashboard,” and achieved a collective submission rate of zero. An emergency all-hands was convened.
“We identified the core issue,” said Priya Suresh, 27, a growth hacker serving as the cohort’s self-appointed Director of Grappling Outcomes. Suresh, who has three-eighths of a stripe on her white belt and a Notion page titled My BJJ OKRs Q2 2026, explained that the cohort had been “optimizing for the wrong north star metric.”
“We were measuring drilling reps,” she said. “But drilling reps don’t translate to outcomes without the qualitative signal. We needed to measure taps-per-session, escape velocity from mount, and guard entry conversion rate. So we built a dashboard.”
The dashboard — a 23-panel analytics report with a filter for “belt progression stage” — was presented to Professor Olvera.
“He looked at it for a long time,” Suresh said. “Then he told me to do shrimping for fifteen minutes. That’s when I knew we were misaligned.”
In Week Six, the cohort voted to pivot.
“The gi is legacy technology,” said Kettler, who announced the decision via a video message from his car. “It introduces friction into the system. No-gi is the modern stack. We’re pivoting to no-gi and compressing the remaining sprint timeline.”
Olvera declined to comment on the pivot but was observed, on multiple occasions, sitting in the corner of the mat with his head in his hands.
The pivot did not improve outcomes. Kettler’s field notes from Week Eight contain the entry: Cohort is still getting tapped immediately. New hypothesis: the problem is the partners, not the framework. Exploring option of AI-simulated resistance.
The AI-simulated resistance, according to Reinholdt, was a grappling dummy with a Bluetooth speaker inside playing an audio clip of Kettler’s voice saying “that’s a tap” at five-second intervals.
“I submitted it twice,” Reinholdt said. “But it didn’t tap back. So I don’t know if those count.”
Demo Day is scheduled for Saturday morning at Cali Ground Warfare. Fourteen investors have RSVP’d, including Wes Chandler, general partner at Meridian Ventures, who told this reporter that BlueSprint Labs “represents a compelling wedge into the seven-billion-dollar global martial arts market.”
Chandler said he expects to see “live demonstrations.”
“Brayden said the cohort would be showing their submissions,” Chandler said. “I’m excited to see what they’ve built.”
Kettler said in a statement that Demo Day would “tell the BlueSprint story through product, not slides.” He confirmed the day’s agenda: a keynote, a 38-slide deck, a 12-minute video compilation of drilling footage set to original hip-hop production, and — time permitting — “a live rolling segment.”
The cohort, reached for comment via the group Slack, had mixed feelings about the live rolling segment.
“I’ve asked Brayden if we can gate that behind a Q&A,” Suresh said.
Reinholdt, asked whether the program had taught him anything, paused for a long time.
“I learned that you can’t A/B test a hip bump sweep,” he said. “Olvera showed me that. Made me do it about eight hundred times. And I still can’t hit it.”
He paused again.
“I might re-enroll in the spring cohort.”
Olvera has requested that investors sign a waiver.
The 12-week sprint ends Friday.