Research Program Manager - Model Development

Full Job Description

Our Mission

Reflection is a research lab making intelligence open and accessible for everyone to use, customize, and build on. We build open models that let anyone control their intelligence and help shape the future of AI. Our mission: make intelligence open and accessible to all.

About the Role

Research Program Managers at Reflection are high-leverage leaders and operators who embed directly with research and engineering teams to accelerate the pace of frontier model development. They are not project trackers. They are force multipliers who bring clarity to ambiguity, drive decisions when the path forward is unclear, and ensure that the work happening across multiple teams connects into a coherent whole.

This role is embedded directly within our Pre-training ML and Data teams, working alongside our Pre-training research leads at the core of how we build frontier models. You will be deeply involved in the research lifecycle, from data pipeline coordination and experiment planning to model architecture decisions and scaling strategies. Your job is to understand how our researchers work, identify the highest-leverage gaps, and build the programs, processes, and coordination structures that let them focus on pushing the frontier rather than navigating organizational complexity.

You bring a first-responder mentality. When things go sideways, you don't wait to be asked. You jump in, assess the situation, cut through noise, align the people who need to be aligned, and drive resolution.

What You'll Do

• Embed within the pre-training ML and Data teams to deeply understand the technical landscape, build trust with researchers and technical leads, and identify where programs and process can have the most impact on research velocity.

• Drive end-to-end execution of complex, cross-team research initiatives that span data, model architecture, training runs, and evaluation, often without established playbooks.

• Coordinate the operational rhythm of pre-training research, including experiment prioritization, run scheduling, data readiness, and checkpoint handoffs to downstream teams.

• Equip research leadership to make decisions quickly by going deep on technical tradeoffs and presenting clear, actionable recommendations rather than status updates.

• Build lightweight processes that bring structure to unstructured research environments without adding friction. Create coordination mechanisms for cross-team handoffs, config management, and research milestones that replace ad hoc Slack threads with durable, visible systems.

• Act as the connective tissue between pre-training, mid-training, post-training, and infrastructure teams, ensuring that upstream decisions propagate cleanly and downstream teams are never surprised.

About You

• 7+ years of experience in technical program management, research operations, or ML engineering, with a track record of building programs from scratch in research or ML-heavy environments.

• Deep enough technically to engage with researchers on topics like model architecture, scaling laws, data processing pipelines, training dynamics, and optimizer behavior. You don't need to run the experiments yourself, but you need to follow the reasoning and spot risks early in order to discuss tradeoffs.

• Proven ability to embed within technical teams and earn trust through competence, reliability, and genuine curiosity about the work. Researchers want to work with you because you make them faster, not because the process says they have to.

• Resourceful and high-agency. You navigate ambiguity and shifting priorities without losing momentum. You figure out what needs to happen and you make it happen.

• Strong stakeholder management skills with the ability to influence senior technical staff through consistent delivery and well-informed judgment, not authority.

• Comfortable in high-stakes environments where decisions impact model quality, compute spend, and training timelines measured in weeks or months.

• Excited to build from zero to one. We are a small, fast-moving team and this role will help define how research program management works at Reflection.

• Motivated by enabling researchers and engineers to build the world's most capable open-weight AI systems.

What We Offer:

We believe that to make intelligence open and accessible to all, you need to start at the foundation. Joining Reflection means building from the ground up as part of a talent-dense team. You will help define our future as a company, and help define the future of open foundational models.

We want you to do the most impactful work of your career with the confidence that you and the people you care about most are supported.

• Top-tier compensation: Salary and equity structured to recognize and retain our talent globally.

• Stock options: Everyone who joins and contributes to Reflection's success gets to share in the upside through stock options.

• Health & wellness: Comprehensive medical, dental, vision, and life, with an annual wellness allowance.

• Meals: Lunch and dinner are provided in the office daily.

• Life & family: 22 weeks paid parental leave for all new birthing and non-birthing parents, including adoptive and surrogate journeys.

• Vacation days: Unlimited paid time off in the U.S. and 30 days in the U.K.

• Sponsorship support: We sponsor visas to help exceptional talent join our team and support long-term immigration pathways where applicable.

• Team building: We have regular off-sites, happy hours, and team celebrations.

Export Control Notice: This position may require access to technology or source code subject to the U.S. Export Administration Regulations. Any offer of employment for this role may be conditioned on the Company's ability to provide the candidate with access to such technology or source code in compliance with applicable U.S. export control laws, which may require the Company to seek government authorization.

Discovered directly from the employer’s public Ashby Job Postings API. The complete public role content and compensation metadata were normalized into safe candidate-facing sections.