AI Research Lead

1mind • San Francisco, CA

7d • Onsite

About The Position

1mind is a platform that deploys multimodal Superhumans for revenue teams. These Superhumans combine a face, a voice, and a GTM brain — equipped with deep technical and product knowledge. They can lead unlimited, simultaneous conversations 24/7, meeting buyers when they’re most active and engaged. Superhumans qualify leads, book meetings, deliver pitches, give interactive demos, handle objections, uncover pain points, build value models, provide support, and onboard customers. They live across websites, inside your product, can join live calls as active participants, and work alongside your team in deal rooms. 1mind Superhumans integrate seamlessly into existing workflows, scale instantly, and drive measurable impact — growing revenue, reducing headcount, accelerating pipeline to closed-won, and creating a more delightful buyer experience.
We’re looking for an AI Research Lead to define and drive 1mind’s research agenda. This is one of the most important hires we’ll make — and the person who fills it will have an outsized impact on the trajectory of the company.
You’ll lead exploratory research into vertical post-training for sales and GTM domains, developing models that understand how humans sell, buy, and build relationships. You’ll work directly with the CTO and have the freedom to shape the research direction, build your own team, and publish your work. This is applied research with active signal: our agents are live in customer environments today, generating high-fidelity data from thousands of real buyer interactions.
If you’ve been doing cutting edge research, post-training work at a frontier lab and want to build something that ships into production on a unique, dataset that no one else has — this is your role.

Requirements

• 4+ years of experience in machine learning research or applied AI, with at least 1–2 years focused on post-training (RLHF, DPO, reward modeling, alignment, or related techniques).
• Deep technical fluency in LLM training pipelines, fine-tuning methodologies, and evaluation frameworks.
• Demonstrated ability to take research from exploration to production-grade systems.
• Strong product intuition — ability to identify where research creates real business value and prioritize accordingly.
• Based in or willing to relocate to San Francisco.

Nice To Haves

• 6+ years of total experience; Staff Researcher/level or equivalent.
• Experience at a frontier research lab
• Experience leading research teams.
• Familiarity with reinforcement learning from real-world feedback loops (not just simulated environments).

Responsibilities

• Own and drive the post-training research roadmap for 1mind’s vertical AI models, from exploration through production deployment.
• Design and execute experiments on sales LLM fine-tuning, copilot behavior modeling, and domain-specific reinforcement learning.
• Leverage 1mind’s live RL environment and high-fidelity reward signals from real-world agent interactions to train and iterate on models.
• Develop novel post-training techniques — RLHF, DPO, reward modeling, and beyond — tailored to GTM and conversational commerce use cases.
• Collaborate cross-functionally with engineering, product, and GTM teams to translate research into measurable product improvements.
• Build and lead the research org over time — hiring, mentoring, and setting the technical bar for a world-class applied research team.
• Evaluate new model architectures, training strategies, and inference optimizations for 1mind’s multimodal agent stack.
• Publish research findings and contribute to open-source and open-weight model initiatives where appropriate.

Benefits

• Build post-training models no one else can. 1mind is the only company with the vertical GTM data and live agent interactions needed to train domain-specific models. You won’t be fine-tuning on synthetic benchmarks — you’ll be training on real sales conversations with real reward signals.
• Live RL environment from day one. Our Superhumans are already operating in the wild, generating detailed reward data from thousands of real buyer interactions. You’ll have a production feedback loop most researchers only dream about.
• Freedom to build. Define the research agenda, choose the problems, hire your team, and shape the direction of a category-defining company.
• Publishing and open source encouraged. We support publishing your work and contributing open-weight models. IP is evaluated case by case, but the default is openness.
• Competitive compensation. We offer aggressive, market-leading compensation for this role, including base salary, equity, and full benefits.
• High-impact, early-stage opportunity. Work directly with a world-class team at a Series A company backed by top investors, with 50+ enterprise customers like LinkedIn, HubSpot, Nutanix, Samsara, and Boston Dynamics.

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level
• Education Level: No Education Listed