MTS: AI Research Engineer
$175k - $225k • London • Full-time
Posted 7d ago
Remote Work Policy
On-site
Employment Type
Full-time
Categories
AI Agent Engineer
About the job
We are seeking an early Member of Technical Staff within our Research Tribe to shape the core systems powering our AI and Agent Systems. This deeply technical role bridges distributed systems, machine intelligence, and product, focusing on turning stochastic models into reliable, production-grade systems. You will design and implement agentic systems at scale, defining agent topologies, orchestration layers, state management, memory, and tool-use execution. You will also develop behavioral and persona modeling, design evaluation frameworks, build signal and insight systems from agent outputs, and engineer production reliability infrastructure for agents.
This position emphasizes building robust, inspectable, and improvable systems over simple prompt chaining. You will work in a fast-paced, iterative environment where research and product development move together, shipping quickly and learning from real users. We value high ownership, low ego, and a comfort with ambiguous problems at the frontier of AI.
Responsibilities
- Design and implement agentic systems at scale, including agent topologies, orchestration layers, state management, memory, and tool-use execution.
- Develop behavioral and persona modeling by simulating user archetypes with goal-directed behavior and designing data-grounded personas.
- Design and build frameworks for evaluation and prompt engineering, measuring behavioral fidelity and failure modes.
- Construct systems to transform raw agent outputs into high-signal insights, including clustering observations and implementing severity scoring.
- Build infrastructure for production reliability of agents, including tracing, attribution, cost monitoring, failure detection, and fallback mechanisms.
Requirements
- 5+ years of experience building complex software.
- 2+ years of experience shipping agentic LLM systems in production environments.
- Experience with multi-agent orchestration, sandboxed execution, and complex backends.
- Expertise in Python, including async programming and structured data modeling (Pydantic).
- Prioritization of state machines, orchestration layers, and tool-use contracts over prompt hacking.
- Ability to build for reliability, inspectability, and improvability.
- Comfortable shipping quickly and refining systems in tight loops.
- High ownership and low ego, with a willingness to be wrong and change one's mind.
- Excitement for ambiguous problems and defining best practices at the frontier of AI.
- Product intuition, prioritizing signal over noise and ensuring built systems help users make decisions.
Benefits
- Salary: $180-250k base
- Equity
- 100% paid medical, dental & vision (US only)