Machine Learning Engineer

Hyperbound builds the Revenue Activation Platform, an agentic operating system for sales that goes beyond recording what happens on calls. This Machine Learning Engineer role owns models end to end, from training and fine-tuning through deployment, and then keeps those models working reliably in production. The work is centered on fine-tuning and deploying open source models, pushing models on-device when needed, and creating evaluation frameworks to verify improvements.

What you’ll do

• Own your models end to end, taking them from training through production.
• Fine-tune and run open source models in production.
• Push models on-device when a customer’s latency or privacy requirements call for it.
• Build evaluation frameworks, benchmarks, and regression suites to determine whether changes actually improve outcomes.
• Build and ship models that power Hyperbound’s roleplay, scoring, and coaching products, including full lifecycle ownership from training and fine-tuning through production deployment and ongoing reliability.
• Fine-tune and deploy open source models where they provide more control over cost, latency, or model capabilities.
• Work closely with the founders and the broader engineering team with real input into what you build next.

How the role fits the work

The team operates on a practical production focus. In this role, you spend most of your time fine-tuning and running open source models in production.

On-site location and schedule

This position is based in San Francisco, CA and is onsite. The team is in the office five days a week.

Compensation and benefits

• Comp: $260,000-$300,000+ per year based on experience, plus meaningful equity.
• 401k
• Medical, dental, vision
• Commuter and parking benefits
• Unlimited PTO
• Free lunch and dinner in the office

Interview process

The process moves quickly: an intro call, a technical conversation with the team you would work with, and a final conversation with the founders. From the first conversation to offer, the timeline is 1 to 2 weeks.