Staff Engineer - Backend

Summary

Join interface.ai, a leading AI provider for banks and credit unions, as a Staff Backend Engineer – Core AI Platform. Architect and lead the development of a foundational multi-agent infrastructure for next-generation intelligent systems in financial institutions. Design and scale custom AI orchestration frameworks integrating language models, memory, judgment modules, and tool use into autonomous systems. Work at the intersection of machine learning, distributed systems, and agentic reasoning, collaborating with researchers, engineers, and product leaders. This role involves building the AI runtime and execution architecture for mission-critical agents in a high-regulation industry. You will define and build the core AI infrastructure powering autonomous financial workflows.

Requirements

• 8+ years of backend or platform engineering experience, with at least 2 years building or deploying AI/ML systems
• Prior hands-on experience building ML models or training pipelines —you know how models learn, behave, and break
• Expert knowledge of LLM system design, agent architectures, and Reinforcement Learning techniques
• Deep experience with Node.js, asynchronous architecture, and performance-critical backend systems
• Proven track record building distributed, event-driven systems in high-throughput environments
• Experience building real-time inference systems that integrate LLMs with retrieval, memory, or tool use
• Strong systems design skills: designing modular, fault-tolerant, observable software at scale
• Demonstrated ability to lead architectural design and cross-functional engineering initiatives

Responsibilities

• Architect a multi-agent AI framework for orchestrating LLMs, tools, memory, and decision modules in live user-facing systems
• Build and optimize low-latency, distributed inference systems that meet real-time SLAs for transactional environments
• Develop modular components for task planning, reward routing, fallback handling, and multi-turn reasoning
• Design developer-facing APIs and tooling to allow AI product teams to safely extend and compose agentic functionality
• Integrate vector stores, custom retrieval pipelines, model evaluators, judgment layers, and auto-tuning workflows
• Drive the implementation of prompt tuning, reward modeling, and LLM-as-a-Judge techniques in production loops
• Collaborate with research to productionize new RL, planning, or alignment strategies
• Establish architectural best practices for extensibility, observability, and trust in AI-enabled systems

Preferred Qualifications

• Experience building multi-agent systems or intelligent orchestration engines
• Familiarity with vector databases, semantic search, and prompt engineering techniques
• Comfort integrating ML eval frameworks and offline/online experimentation pipelines
• Open-source contributions to LLM tooling or infrastructure a plus