Description
We're hiring Senior ML Engineers to own major components end-to-end for our Agent Platform, a production-grade platform for autonomous AI agents. You'll work at the intersection of applied ML and systems engineering, shaping a system that serves Fortune 500 marketing teams.
Responsibilities:
- Build core agent infrastructure, including the agent runtime, tool execution layer, memory systems, sandboxed execution, or control plane
- Design ML workflows at enterprise scale, including model customization, serving, and lifecycle management
- Innovate and explore new approaches to agent reasoning, tool orchestration, memory, or evaluation
- Close the loop with customers, joining regular customer engagements to see how your systems perform in real deployments
- Own what you ship, from architecture through production operations, including deployment, monitoring, observability, and incident response
Requirements:
- Ph.D. or M.S. in Computer Science or related field
- 5+ years of experience building and deploying production ML systems
- Strong software engineering fundamentals, including proficiency in Python and/or Java
- Deep hands-on experience with at least one modern deep learning framework (PyTorch, TensorFlow, JAX)
- Production experience with LLMs, including prompt/context engineering, working with LLM APIs, fine-tuning, or building LLM-powered applications
- Experience with cloud platforms (AWS or Azure) and data infrastructure (Postgres, Redis, Elasticsearch, Snowflake, or similar)
Nice to Have:
- Agent or LLM infrastructure depth, including building agent loops, tool-use orchestration, RAG pipelines, long-term memory systems, or fine-tuning/serving infrastructure
- Platform-scale systems thinking, including designing catalog systems, plugin architectures, or intent routing
- ML-Ops or Agent-Ops experience, including building eval frameworks, execution tracing, drift detection, guardrails, or HITL intervention systems
- Multiplier instincts, including mentoring engineers, shaping a team's technical roadmap, or building internal tools and practices