See how candidates think and approach the work this role demands, before the phone screen. We'll build a video challenge from this posting, and you can edit or share it before it goes live.
A frontier open-model lab is standing up a cleared, on-site deployment function to put its models inside classified and air-gapped government networks, signaling a deliberate push into national-security customers rather than only API and enterprise sales.
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.
Forward Deployed Engineers are Reflection’s technical owners in every customer engagement. You will own the technical execution of strategic engagements end to end, from discovery and solution architecture through proof-of-concept and production delivery. The work spans every layer of the AI stack: GPU orchestration, inference, model post-training, and the custom harnesses and applications built on top.
We’re seeking exceptional people with depth across the AI stack who can partner with customer’s ML and engineering teams, win and deliver strategic engagements, and turn what they learn in the field into input that shapes our models and platform.
In this role, you will support U.S. government customers, deploying Reflection’s open models into secure, classified, or edge environments.
This position requires an active U.S. government clearance (TS/SCI preferred).
Secure Deployment: Design, deploy, and operate agentic systems in air-gapped and classified networks, including fully offline model serving without access to external services.
Mission Engagement: Work on-site with government customers, including in SCIFs. Turn mission requirements into secure AI solutions that meet accreditation and compliance standards.
Technical Discovery & Architecture: Lead deep technical discovery with customer teams across inference workloads, agentic applications and post-training. Understand customer needs and translate them into feasible technical solutions with clear success criteria to make Reflection the technical winner in every deal.
Benchmarks & POCs: Own proofs of concept end to end: define evaluation criteria with the customer, benchmark on their real workloads, and demonstrate Reflection’s quality, latency, and cost against competing options.
Delivery: Architect and build complex AI systems, including agentic applications, model-serving infrastructure, evaluations, and post-training recipes. Own the customer engagement and relationship: define project plans, sequence and manage work across workstreams, and keep customer stakeholders aligned.
Product & Research Contribution: Identify repeatable patterns and model gaps in the field that should shape our platform or model capabilities. Build prototypes, validate that they meet customer needs, and generalize them into Reflection’s platform and research priorities.
Playbooks & Operating Model: Help shape and scale the deployment organization by defining playbooks, reference architectures, reusable technical assets and best practices that make every teammate more effective and support the team’s growth.
Experience delivering software into classified or air-gapped government environments: you know the security controls, the accreditation processes, and the limits on offline deployment.
Strong technical fluency: comfortable reading, writing and reasoning about code, architecture and system behavior, with strong Python and experience shipping production-grade systems.
A track record of leading complex technical projects: coordinating contributors across work-streams, making practical tradeoffs, removing obstacles, and owning end-to-end delivery outcomes.
Hands-on infrastructure and inference experience: you’ve deployed multi-node GPU clusters (Kubernetes or Slurm), served large models in production (vLLM, SGLang), and rigorously benchmarked tradeoffs like quantization, caching, batching and speculative decoding.
Agent engineering: you’ve shipped LLM-based agents to production in the last 1–2 years, designed their harnesses (tool calling and MCP, context management, orchestration, guardrails) and the evals behind them, and know when post-training can be used.
5+ years of relevant experience delivering technical systems, including 2+ years in a technical leadership role (e.g., tech lead, engineering manager or founding engineer).
Demonstrated ability and interest in customer-facing work: understanding user needs, architecting solutions for real business problems, and delivering tangible outcomes. You can go from a whiteboard session to debugging a deployment in the customer’s environment.
Self-starter with high agency and ownership who excels in fast-paced startup environments where the playbooks are still being written.
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.
Equal Employment Opportunity Statement
Reflection AI is an equal opportunity employer. We do not discriminate on the basis of race, color, religion, sex (including pregnancy), sexual orientation, gender identity or expression, national origin, ancestry, age, marital status, disability, genetic information, veteran or military status, or any other characteristic protected by applicable law. We encourage applicants from all backgrounds to apply.
We provide reasonable accommodations throughout the hiring process for applicants with disabilities, pregnancy-related limitations, or religious needs. If you need an accommodation, contact us at accommodations@reflection.ai.
We will consider for employment qualified applicants with arrest and conviction records in accordance with applicable law, including the San Francisco Fair Chance Ordinance and the California Fair Chance Act.
Candidate and Employee Privacy Notice: We collect and process personal data about applicants for the purposes described in our Candidate and Employee Privacy Notice
Answer a McCoy challenge. Companies hiring in engineering see your response and reach out when they want to talk.
About 15 minutes to prepare and 5 minutes to record. Unlimited retakes and a teleprompter help you answer. No AI training, and you can delete your response anytime.
This isn't an application to Reflection AI.
reflection-ai:dea72756-5e23-4a04-813f-5af718605edbSee how we measure for definitions, or our corrections log for known issues. Found something wrong? Flag a correction.
reflectionaiThis posting hasn't been probed by our closure verifier yet. Stream C runs on a rolling schedule against postings approaching the close-decision threshold.