
Research Engineer, Agentic Systems
Posted 1 day ago

Posted 1 day ago
This is a fully remote position, open to applicants in United States.
• Design and implement agentic systems that allow thousands of agents to explore extensive search spaces.
• Determine how to decompose problems, what information each agent receives, the number of agents to deploy, and which model should execute each phase.
• Enhance model accuracy by focusing on context design, self-verification, voting, and achieving consensus among agents.
• Assess quality through sampling, automated verification, and measuring precision and recall on final outputs.
• Minimize costs and latency through model routing, caching, early stopping techniques, and utilizing smaller or fine-tuned models.
• Transform one-time experiments into reusable workflows utilizing the current orchestration stack.
• Assess new models, inference configurations, and training methodologies, implementing those that prove effective.
• Manage systems from initial prototype stages through to production deployment.
• Proven experience in developing applied AI systems and LLM applications that have been successfully launched.
• Practical experience with multi-agent systems or high-fan-out LLM workloads: parallel agents, map-reduce applications, and large batch inference.
• Strong programming skills in Python; capable of building concurrent, distributed systems that are reliable in production environments.
• Ability to evaluate output quality: you define the expected outcomes before scaling operations and can distinguish between signal and noise among agents.
• A commitment to measurement: tracking cost, yield, false-positive rates, and assessing the impact of changes made.
• Ownership mentality: capable of managing a roadmap and a production system with minimal supervision.
• Proficient written communication skills.
• Experience in building agent orchestration or job systems (Ray, Temporal, Modal, or custom solutions).
• Familiarity with using LLMs to search extensive datasets: codebases, document collections, logs, or other large corpora.
• Experience with post-training or hosting open-source models for high-volume inference scenarios.
• Background in distributed systems.
• Publicly accessible work that we can review: an agent framework, an open-source project, a technical write-up, or a research paper.
• Flexible PTO.
• Centralized ordering for home-office equipment.
• Health and wellness stipend.
• Budget for travel within the office.
• Weekly coffee stipend.
• Comprehensive medical, dental & vision insurance coverage for you, with partial coverage for dependents.
• Annual membership with One Medical.
• 401(k) plan with employer matching on contributions made while employed at Ramp.
• Fertility HRA providing up to $10,000 annually.
• Parental leave policy: up to 16 weeks (for birthing + bonding) or 8 weeks (bonding only) at full pay.
• Pet insurance options available.
• In-office perks including lunch, snacks, drinks, and more.
• Coverage for relocation expenses to NYC or SF, if required.
• Group medical, dental, and vision coverage provided through Sun Life.
• Life, AD&D, and disability insurance coverage.
• Fertility drug coverage up to a lifetime maximum of $4,000.
• Group Retirement Plan with employer matching (RRSP + DPSP).
• Employee Assistance Program and virtual care services through Lumino Health.
• Private medical insurance through Freedom Elite.
• Access to virtual GP and at-home care via eMed x Livi.
• Workplace pension plan through Penfold, with a salary sacrifice option.
Dragos, Inc.
Dragos, Inc.
Thomson Reuters
Get handpicked remote jobs straight to your inbox weekly.