Responsibilities
- Design and lead the architecture of a distributed, multi-agent AI framework integrating LLMs, decision layers, memory, and tooling.
- Build real-time, low-latency inference systems capable of meeting the SLAs of transactional and regulated environments.
- Develop modular backend components for agent planning, fallback handling, reward routing, and continuous reasoning.
- Create developer-facing APIs and tools enabling safe, composable extensions of AI capabilities.
- Integrate vector databases, custom retrieval mechanisms, evaluators, and auto-tuning pipelines into live systems.
- Work closely with AI researchers to productionize experimental models, RL strategies, and alignment techniques.
- Establish architectural best practices for scalable, fault-tolerant, and policy-aligned AI-driven systems.
- Lead technical direction and influence cross-functional engineering strategy.
Requirements
- 8+ years in backend/platform engineering, including 2+ years building or deploying ML/AI systems in production.
- Strong experience with Node.js and asynchronous, event-driven architecture.
- Expertise in large language model (LLM) systems, reinforcement learning (RL), and agent architecture design.
- Proven record building distributed systems and high-throughput infrastructure with real-time inference capabilities.
- Prior hands-on ML experience—model training, pipeline design, or production debugging.
- Proficient in modular software design with observability, testing, and resiliency in mind.
Nice to Have
- Bonus points for experience with vector databases, prompt engineering, semantic search, or orchestration engines.
- Open-source contributions in the LLM/AI infrastructure space are a plus.
Benefits
- Remote-first work model with flexibility across time zones.
- Comprehensive health insurance to support your well-being.
- Career growth in one of the fastest-growing AI/FinTech environments.
- Opportunity to lead in an emerging field at the intersection of AI, banking, and compliance.
- Dynamic team culture combining AI research, product design, and engineering excellence.
Additional Information
- Position posted by Jobgether on behalf of interface.ai.
- Application process uses AI-powered screening to analyze CV and LinkedIn profile, focusing on skills, experience, and achievements.
- Match score is determined by comparison to job’s core requirements and past success factors.
- Top 3 candidates with highest match score are shortlisted automatically.
- Human review may occur to ensure no strong profile is missed.
- Process is transparent, skills-based, and free of bias.
- Shortlisted candidates are shared directly with the company (interface.ai) for final decision and next steps.
- Final decision and interview process conducted by internal hiring team of the company.
