Software Engineer, Data Infrastructure

Scale AI is seeking a highly skilled and motivated Software Engineer to join our dynamic Public Sector Engineering team. As a part of this team, you will play a critical role in supporting Scale’s government customers by scoping and developing onsite solutions. Our scalable, high-performance platform is the foundation for these customer solutions, and your expertise will be instrumental in designing and implementing systems that can handle interactions with existing customer systems to help our products integrate into existing customer workflows. The Role - We are looking for an exceptional Senior Software Engineer to architect and build the foundational data infrastructure that will serve as the brain of a project ecosystem. - We are not looking for someone to stitch together off-the-shelf data frameworks. You will be responsible for designing highly novel data models and processing pipelines capable of handling massive quantities of output data from complex simulations. - At the core of this role is the challenge of building a foundational data ensemble—a unified architecture that seamlessly aggregates, structures, and stages diverse sources of simulation outputs and user inputs. Your systems will manage enormous batch throughput jobs with strict, minimal latency requirements, ensuring that downstream AI systems and language models have the exact context they need to actionably reason over complex, multi-dimensional scenarios. Key Responsibilities - Architect the Data Ensemble: Design and implement the architecture to ensemble various sources of injected context (deeply structural simulation data, historical game states, and dynamic user inputs) into a unified, highly queryable format optimized for LLM consumption. - Massive Batch Infrastructure: Build highly scalable, resilient data architectures from scratch. You will optimize for moving, transforming, and processing massive quantities of simulation output data via enormous batch jobs, maintaining the minimal latency required for rapid wargame iterations. - Complex Data Modeling: Design sophisticated, highly relational data models that accurately represent massive, state-based simulation environments, making them easily interpretable by machine learning models. - First-Principles Problem Solving: Navigate highly ambiguous product requirements to design custom, ground-up systems where existing open-source or enterprise tools simply cannot handle the structural complexity or scale. - Technical Leadership: Set the technical standard for the data infrastructure team, driving rigorous code quality, system performance, and architectural clarity. What We’re Looking For - Experience: 5+ years of backend or data infrastructure experience, operating at a Senior, Staff, or Principal level. - Engineering Excellence: Deep, expert-level proficiency in systems languages (e.g., Rust, Go, C++, or highly optimized Python/Java, Spark, PySpark) and a fundamental understanding of memory management, compute limits, and distributed systems architecture. - High-Throughput / Low-Latency Data: Proven track record of processing massive datasets. You understand how to optimize massive batch jobs and parallel processing across distributed simulation nodes without sacrificing speed. - Information Retrieval & Context Surfacing: You don't need a background in AI agents, but you must be an expert in surfacing the right needle from an ocean of hay to feed decision-making engines. We highly value engineers with backgrounds in: - Search & RecSys: Building complex information retrieval systems or recommendation engines. - Gaming / MMOs: Managing complex state, data relationships, and telemetry for massive, highly populated simulations. - High-Frequency Trading (HFT): Processing disparate, massive streams of data for algorithmic decision-making. - Mission-Driven: A strong desire to build robust, foundational technology that supports national security and defense modernization. Nice to Have - - Security Clearance: An active Secret or TS/SCI clearance is a nice to have for this role. If you do not have an active clearance, you must be eligible and willing to obtain one. - Experience with LLM context optimization, vector embeddings, or agentic AI frameworks (e.g., advanced RAG architectures). - Deep domain experience working with wargaming data, complex systems modeling, or distributed simulation protocols. - Previous experience in a high-growth, 0-to-1 startup environment. Compensation packages at Scale for eligible roles include base salary, equity, and benefits. The range displayed on each job posting reflects the minimum and maximum target for new hire salaries for the position and may be inclusive of several career levels at Scale; it will be determined during the interview process based on work location and additional factors, including job-related skills, experience, qualifications, interview performance, and relevant education or tra

Experience: 5+ years of backend or data infrastructure experience, operating at a Senior, Staff, or Principal level. Engineering Excellence: deep, expert-level proficiency in systems languages (e.g., Rust, Go, C++, or highly optimized Python/Java, Spark, PySpark) and a fundamental understanding of memory management, compute limits, and distributed systems architecture. High-Throughput / Low-Latency Data: proven track record of processing massive datasets with optimized batch jobs and parallel processing across distributed nodes without sacrificing speed. Information Retrieval & Context Surfacing: expertise in surfacing the right data to feed decision-making engines. Nice to Have: Security Clearance (active Secret or TS/SCI) or eligibility to obtain one; experience with LLM context optimization, vector embeddings, or agentic AI frameworks (advanced RAG architectures); deep domain experience with wargaming data, distributed simulation protocols; experience in 0-to-1 high-growth startup environments.

Architect the Data Ensemble: design and implement architecture to ensemble various sources of injected context into a unified, highly queryable format optimized for LLM consumption. Massive Batch Infrastructure: build scalable data architectures from scratch to move, transform, and process massive quantities of simulation output data with minimal latency. Complex Data Modeling: design sophisticated, highly relational data models representing massive state-based simulation environments. First-Principles Problem Solving: design custom, ground-up systems where existing tools cannot handle the complexity or scale. Technical Leadership: set the technical standard for the data infrastructure team, driving code quality, system performance, and architectural clarity.