We compress intelligence into a milliwatt.

Fermion builds small, fast AI models that run on people's own devices and on companies' own servers. Phonon-2, our speech model, has been the #1 trending speech recognition model on Hugging Face since launch and is already used inside more than 50 companies. Our launches have reached over 1.5 million people on X. Detta, our Mac dictation app, passed 10,000 users in its first four days.

This role owns growth and sales across Detta, enterprise deployments and the consumer products we launch next.

Detta

Detta is dictation for the Mac: hold a key, speak, and the text appears in whatever app you are typing in. Recognition runs on the Mac itself, so it is fast, works offline and nothing leaves the device. It competes with apps like Wispr Flow and Superwhisper.

You will own Detta's growth, for individual users and for companies.

Individual users

Organic acquisition: find the channels that work, such as X, YouTube, Reddit and productivity communities, Product Hunt, and comparison and search pages. Run them and measure each one.
Paid acquisition: run and optimise paid campaigns on Meta, Google, YouTube, Reddit and X, plus sponsored creator and newsletter placements. Own budgets, creative testing, landing pages and attribution, and scale what pays back.
Content and SEO: comparison pages, guides, demo videos and posts that bring people to Detta through search and social.
Activation: get every new user from download to their first successful dictation, and from there to using Detta every day. Work with engineering on onboarding, permissions and the first-run experience.
Retention: understand who keeps using Detta and who drops off. Talk to users every week and run lifecycle email.
Referrals: get users to bring their friends and teammates.
Pricing: help design and test pricing and a paid plan.
Launches: plan and run each Detta release.

Detta for companies

Because Detta runs entirely on the device, voice and text never leave the company's machines. That matters to legal, healthcare, finance and any team with sensitive data, and it is a big part of why companies ask about rolling it out.

Sales: find the teams where dictation saves the most time, such as support, sales, legal, healthcare and engineering. Run pilots and expand them into company-wide rollouts.
Team plan: work with us on what companies need: volume pricing, central billing, managed installs across company Macs, and shared custom vocabulary for product names, people and internal terms.
Compliance: find out early what each company requires, such as SOC 2 or a HIPAA BAA, and work with us to meet it.

Enterprise deployments

Companies are paying a lot for general-purpose AI APIs that are slower and more expensive than they need to be. We build custom models trained on a company's own data, sized for the job and deployed on their cloud, their own servers or their devices. That means speech recognition for transcription-heavy products, fast decision and classification models for routing, triage and moderation, and small language models that run on-device. The customer pays for model development once, then for inference on an ongoing basis.

Pipeline: define who we sell to and build the target list. Look at companies already using our open models, YC and other startups, and larger teams with heavy AI bills.
Outbound: reach those teams through cold email, LinkedIn, warm introductions and events. Write the messaging.
Inbound: follow up with the teams already running Phonon-2.
Discovery and scoping: run first calls, understand the customer's problem, data and hardware, and bring in our research team to scope a pilot.
Closing: write proposals, handle pricing conversations and get pilots signed. Then turn successful pilots into long-term inference contracts.
Sales material: one-pagers, demos and benchmarks that show what a custom model saves.
Process: set up and run the CRM.

New consumer products

We plan to launch more consumer apps built on our models. You will help decide what we build and lead the launches.

About you

You have grown a consumer or developer product and can show the numbers you moved.
You have led an enterprise sales team, or been a major part of one, ideally selling software to technical teams.
You understand inference economics: what it costs to serve a model, how API pricing compares with running your own, and where the savings come from. If you don't yet, you can learn it quickly.
You have run paid acquisition.
You write clearly and can explain technical work without overselling it.
You know how launches spread on X, Hacker News and Reddit.
You are comfortable owning both a consumer funnel and a sales pipeline at the same time.

Bonus: experience selling AI or developer tools, video or design skills, a technical background.

What we offer

Founding equity and a competitive salary.
Ownership of growth and sales from the start.
Fly to any customer or prospect, anytime, without asking for approval.
Top-end hardware and tools for whatever the role requires.

Apply

Email careers@fermionresearch.com with links to things you have grown, launched or sold, and a short note on what you would try first for Detta and for enterprise sales.

Founding Growth Lead