Software Engineer - Backend

Company Overview Realm builds AI for the mining industry. We integrate with existing mine infrastructure — cameras, SCADA systems, drones, and data repositories — to predict equipment failures, detect unsafe behaviors, and optimize operations. Our platform deploys intelligence in various ways to improve productivity and safety — monitoring PPE compliance via computer vision, optimizing froth flotation processes, analyzing longwall shearer performance, tracking MSHA citation patterns, and more. Our founders started their careers at Palantir before going on to build Blend Labs (publicly traded fintech, founded as CTO) and Even (acquired by Walmart). Our leadership team brings experience from Uber, Metropolis, and other high-bar organizations. We're a small, senior team that ships daily to production mines. AI isn't just our product — it's how we build. We're an AI-first engineering team that uses the latest models and tooling throughout our development process, moving faster and at higher quality than traditional teams our size. Our culture values high agency, deep ownership, collaboration, and autonomy across a fully remote workforce. Teams here move quickly and make meaningful contributions to a product that shapes global industrial outcomes. What You’ll Do Build and extend our TypeScript backend serving real-time mining operations data across web and mobile clients Work across the stack: backend services (50%), applied AI/ML (25%), React frontend (15%), and infrastructure (10%) Build computer vision pipelines — integrating services like AWS Rekognition, TensorFlow, and SAM3 for PPE detection, person recognition, and safety monitoring Integrate with industrial systems (SCADA/OPC-UA) to monitor and optimize mine equipment in real time Ship features end-to-end Deploy to production multiple times per day with CI/CD Work with LLMs for intelligent analysis and automation features Own features in a small team where your code runs at active mines Interact directly with users to deeply understand customer needs. Translate real-world operational challenges into intuitive, high-impact features Work closely with our CEO — former CTO and cofounder of Blend Labs — to build a world-class engineering organization from the ground up. Tech Stack Backend: TypeScript, Prisma, PostgreSQL, ClickHouse, Rust, Python, GraphQL Frontend: TypeScript, React AI/ML and Computer Vision: LLMs, TensorFlow, SAM3, OpenCV, Rekognition Infrastructure: Docker, Kubernetes, DigitalOcean, AWS What We’re Looking For Must-Have Qualifications Experience & Skills 2+ years of professional software engineering experience (high-quality internships included) Strong coding skills, problem-solving ability, and CS fundamentals (data structures, algorithms, system design) Proficiency with agentic coding tools (Claude Code, Codex, Cursor) and AI-first development style Comfort working across the stack — you don't need to be an expert in everything, but you're willing to write a React component, debug a Kubernetes pod, or tune a database query as needed Engineering Judgment & Quality Great judgment and intuition about security, testing, code quality, and tech debt Appreciation of continuous delivery, clean code, and automated testing Ability to design good generalizations while avoiding premature overengineering Ownership & Independence Entrepreneurial work ethic, drive for excellence, and belief in meritocracy, personal responsibility, and hard work Ability to own features end-to-end with minimal direction Highly independent and able to come up to speed without heavy assistance Work Style & Collaboration Experience and comfort working on a remote team Ability to handle multiple work streams and frequent context switching as priorities change Flexible work hours (occasional nights and weekends) when necessary to sprint for urgent deliverables Humility to learn from others while also being able to advocate for and execute on improvements you believe in Eligibility US citizen or permanent resident, and based in the US Nice-to-Have Qualifications Startup experience — comfort with ambiguity and wearing multiple hats Strong TypeScript proficiency Strong network of top-tier engineers Familiarity with computer vision or ML inference pipelines Experience integrating with industrial/IoT systems Functional programming style preferences (immutability, declarative patterns) Interview Process Engineering quiz (async, ~60 min) — A 10-question code reading and analysis assessment. You pick your language (TypeScript, Python, Go, or C++) and answer questions covering code comprehension, bug detection, performance analysis, code review, and system design. No trick questions — we're testing how you think about real code. Live coding interviews (~3 sessions) — Algorithmic problem solving and simulated development tasks representative of actual work. Behavioral interview — Culture fit, working style, and past experience. Reference checks