NodJobsNodJobs

Research Engineer, Agentic Systems

Ramp Β· Remote (US)

πŸ‡ΊπŸ‡Έ Remote (US)Full-time4h ago

At a glance

Where
Remote β€” open to applicants in the United States
Pay
Not disclosed by Ramp
Posted
November 18, 2026
Last confirmed open
1h ago
At Ramp
14 open remote roles β€” mostly Sales & Business (7), Software Development (3), DevOps & Sysadmin (1). Pay is listed on 2 of them ($60,000 – $159,000/yr).
How to apply
Directly on Ramp's own site β€” no NodJobs account needed.

About the role

This role designs and runs multi-agent systems that fan out work across large search spaces, covering task decomposition, agent coordination, result verification and cost control, for Ramp Labs' finance AI products. It suits engineers with shipped applied AI or LLM application experience and hands-on work with parallel or high-fan-out LLM workloads.

What you'll do

  • Decide how work is broken down, what each agent sees, how many agents run, and which model handles each step
  • Improve accuracy through context design, self-verification, voting and consensus across agents, and judge when more agents stop helping
  • Measure output quality via sampling, automated verification, and precision and recall on final results
  • Reduce cost and latency per useful result through model routing, caching, early stopping, and smaller or fine-tuned models
  • Turn one-off experiments into reusable workflows on the existing orchestration stack
  • Evaluate new models, inference setups and training approaches, and adopt what works
  • Design and run systems for how work gets split, how agents coordinate, and how results are deduplicated, verified and ranked
  • Own problems from prototype to production
  • Evaluate costs of multi-agent systems and manage cost per useful result

What you'll need

  • Experience building applied AI systems and LLM applications that have shipped
  • Hands-on experience with multi-agent or high-fan-out LLM workloads: parallel agents, map-reduce over models, large batch inference
  • Strong Python engineering; builds concurrent, distributed systems that hold up in production
  • Judgment about output: decides the desired end result before scaling the swarm and separates signal from agent noise
  • Habit of measuring cost, yield, false-positive rate, and whether a change helped
  • Ownership: can run a roadmap and a production system with little oversight
  • Clear written communication

Nice to have

  • Built agent orchestration or job systems (Ray, Temporal, Modal or custom)
  • Used LLMs to search large spaces such as codebases, document sets, logs or other large corpora
  • Post-training or hosting open-source models for high-volume inference
  • Distributed systems background
  • Public work such as an agent framework, open-source project, write-up or paper

Benefits listed by Ramp

  • Flexible PTO
  • Centralized home-office equipment ordering
  • Health and wellness stipend
  • Budget for intra-office travel
  • Weekly coffee stipend
  • 100% medical, dental and vision coverage for employee, partial for dependents (US)
  • One Medical annual membership (US)
  • 401(k) with employer match (US)
  • Parental leave up to 16 weeks at 100% pay (US, Canada, UK)
  • Pet insurance (US)
  • In-office perks including lunch and snacks (US)
  • Relocation expense coverage to NYC or SF if needed (US)

Tools & skills mentioned

Python

Summarized by NodJobs from Ramp's posting. Read the full posting on Ramp's site β†’

NodJobs is free for job seekers β€” we never charge to apply, and this role takes you straight to the company's own application.

Similar live jobs