Note: The job is a remote job and is open to candidates in USA. LightBox is seeking a Senior Engineer - AI Platform to build self-service infrastructure and developer-facing tooling that makes its AI capabilities accessible and reliable. The role involves developing API gateways, model routing layers, integration surfaces, observability infrastructure, retrieval and caching systems, and model-serving capabilities across internal and external applications.
Responsibilities
- Design and implement API gateways, model routing layers, and service infrastructure that connect fine-tuned CRE models with production applications
- Develop internal developer tooling - prompt management interfaces, evaluation dashboards, trace viewers - that accelerate the team's ability to ship and iterate on AI features
- Implement integration patterns for embedding AI capabilities into LightBox's existing workflow applications, ensuring seamless user experiences
- Build observability and monitoring infrastructure for AI workloads: latency tracking, quality metrics, cost attribution, and usage analytics
- Contribute to the platform's knowledge retrieval and caching layers, optimizing for both accuracy and performance at scale
- Collaborate with ML engineers on model serving infrastructure, helping bridge the gap between training outputs and production deployment
- Participate in architecture decisions and technical design reviews, bringing a pragmatic full-stack perspective to the team's work
Skills
- 6+ years of software engineering experience with a strong full-stack foundation (backend-leaning preferred)
- Production experience building APIs, microservices, and platform infrastructure in Python and/or TypeScript/Node.js
- Solid understanding of cloud infrastructure (AWS or GCP), containerization (Docker, Kubernetes), and CI/CD pipelines
- Experience building developer-facing tools, SDKs, or internal platforms that other engineers consume
- Familiarity with LLM integration patterns: API-based model consumption, prompt engineering, retrieval-augmented generation, and tool/function calling
- Strong fundamentals in distributed systems, data modeling, and API design
- Comfort working in a small team with high autonomy - you can own a feature end-to-end from design through deployment and monitoring
- Experience with Model Context Protocol (MCP) or similar structured AI integration standards
- Background in data-intensive platforms, particularly property data, geospatial systems, or document processing pipelines
- Familiarity with model serving frameworks (LiteLLM, OpenRoute, AWS BedRock) and inference optimization
- Experience in real estate technology, financial services, or other regulated/document-heavy industries
- Prior work building observability or evaluation tooling for ML/AI systems
Company Overview
Company H1B Sponsorship