Middle AI Engineer (LLMs & Spatial AI)

The Purpose

Higharc is building an AI platform that transforms residential construction by automating the design, permitting, and documentation workflows that currently consume months of manual work and create bottlenecks across the homebuilding industry. This Research Engineer role exists to develop foundational spatial AI capabilities—combining reinforcement learning, computer vision, and domain-specific language models—that enable the platform to understand architectural constraints, generate compliant designs, and ground AI reasoning in real building code and construction knowledge, directly reducing design cycles and accelerating home delivery.

What You'll Do

• Design, build, and operate AI-powered features and systems aligned with LLMs, RAG & Spatial AI, delivering intelligent, production-grade product experiences.
• Develop and integrate AI capabilities such as LLMs, agents, embeddings, and inference pipelines into backend services and applications.
• Architect AI workflows including prompt orchestration, tool calling, retrieval augmentation (RAG), and model routing.
• Own the end-to-end lifecycle of AI systems — from experimentation and prototyping to deployment, monitoring, and continuous improvement.
• Build scalable APIs and services that expose AI functionality reliably and securely to product surfaces.
• Collaborate closely with product, backend, and design teams to translate business problems into effective AI-driven solutions.
• Ensure AI system quality through evaluation, testing, observability, and performance monitoring.
• Address AI-specific concerns including latency, cost optimization, reliability, hallucination mitigation, and safety constraints.
• Stay current with emerging AI models, frameworks, and best practices, applying them pragmatically to real-world systems.
• Design and integrate AI capabilities aligned with LLMs, RAG & Spatial AI into product features and backend systems.
• Build and maintain AI workflows such as prompt orchestration, retrieval augmentation, and agent-based execution to solve real product problems.
• Design and conduct original research in spatial reasoning for residential construction, developing models that understand architectural structure, geometric constraints, and spatial relationships between building components.
• Define construction-domain benchmarks, build ground-truth datasets, and establish evaluation pipelines measuring model performance against real homebuilding tasks.

What We're Looking For

• 5+ years of professional software engineering experience, with at least 3 years building AI-powered systems in production.
• Strong experience integrating LLMs, ML models, or AI services into backend systems and applications.
• Solid understanding of AI system design patterns, including RAG, agents, prompt engineering, and orchestration.
• Experience designing and consuming APIs for AI-powered services.
• Hands-on experience evaluating model quality, performance, and reliability in real-world environments.
• Strong programming skills in at least one backend language (e.g., Python, Node.js, Java, or similar).
• Ability to reason about trade-offs between accuracy, latency, cost, and system complexity.
• Hands-on experience building AI-powered features using modern AI models or frameworks aligned with LLMs, RAG & Spatial AI.
• Solid understanding of AI system design trade-offs, including accuracy, latency, cost, and reliability.
• Hands-on experience with 3D modeling, computational geometry, or computer graphics in a research or production context, combined with computer vision techniques for architectural inputs.
• Proficiency in React/TypeScript for integration work alongside ML pipelines and serving-ready model deployments.

Nice-to-Haves

• Experience training, fine-tuning, or adapting ML models for production use.
• Familiarity with vector databases, embedding strategies, and semantic search.
• Knowledge of AI safety, governance, and responsible AI practices.
• Experience with multi-modal AI systems (text, vision, speech).
• Building Higharc AI Platform .
• Familiarity with BIM tools (Revit, AutoCAD) or BIM data formats (IFC, gbXML) for architectural and construction domain knowledge.
• Experience with Three.js or equivalent 3D frameworks; Ph.D. in ML/AI or open-source contributions and published research.

How You Would Fit

• Strong Ownership – you take responsibility end-to-end, not just for tasks but for outcomes.
• Clear Communication – you share context early, write clearly, and make complex ideas easy for others.
• Problem Solving – you move fast from ambiguity to structure, breaking problems into clear, actionable paths.
• Structural Thinking – you design systems, not hacks. Your work scales and your reasoning is grounded in logic.
• High Velocity – you bias for progress, ship iteratively, and know that momentum compounds.

What We Offer

• Supportive culture – feedback flows freely, the best ideas win, freedom to experiment, learn, and grow.
• Challenging – you'll tackle complex, high-impact work that sharpens your skills and accelerates growth.
• Career growth – long-term, stable job with real progression opportunities.
• Competitive salary, benefits, and paid time off.