SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 260,000 - 300,000 / annual
Hyperbound is the Revenue Activation Platform, an agentic operating system for sales founded in 2023 out of Y Combinator (S23). The company is 45 people, Series A, with $18M raised. Last quarter revenue nearly doubled with net revenue retention above 300%. Customers include LinkedIn, Workday, and Intel.
You will own machine learning models end-to-end across Hyperbound's roleplay, scoring, and coaching products—real systems running on actual sales calls. Your responsibilities include:
• Training, fine-tuning, and deploying open-source models to production, with control over cost, latency, and model capabilities
• Getting models running on-device where customer latency or privacy requirements demand it, managing tradeoffs around quantization and distillation
• Building evaluation frameworks, benchmarks, and regression suites to validate whether changes actually improve performance before customers experience them
• Full lifecycle ownership from training through production deployment and ongoing maintenance
• Close collaboration with founders and engineering team, with real input into product direction
The team is assembling builders and operators who move fast and expect high standards. Engineers here are pragmatic—nobody hides behind "that's not my problem," and people fix broken things directly. The company operates five days a week in-office in San Francisco, believing that co-location enables faster iteration and better decision-making. The culture balances high standards with reasonable expectations: people have families, leave for dinner, and return to ship the next day.
Equity is meaningful with real secondary opportunities. The interview process is fast: intro call, technical conversation with your actual team, and final conversation with founders—typically 1-2 weeks from first conversation to offer.
REQUIREMENTS:
The posting does not explicitly state years of experience, required skills, education, or certifications. However, the role implies senior-level capability: end-to-end model ownership, production deployment experience, familiarity with fine-tuning and open-source models, on-device deployment (quantization/distillation), and evaluation framework design are expected.
