Applied AI Research Engineer – ML Systems & Structured Data

Granica•San Francisco, CA

2d•$160,000 - $240,000•Onsite

About The Position

Granica is building the next generation of efficient AI infrastructure. Today’s AI systems are limited not only by model design but by the inefficiency of the data that feeds them. At enterprise scale, redundant data, inefficient representations, and poorly optimized learning pipelines create enormous cost and latency. Granica’s mission is to eliminate that inefficiency. We combine advances in information theory, machine learning, and distributed systems to design data infrastructure that continuously improves how information is represented and used by AI. Granica’s research effort is led by Prof. Andrea Montanari (Stanford) and focuses on building learning systems that operate efficiently on large-scale structured and tabular data. While much of the industry focuses on text or media models, Granica is building the foundations of AI systems that learn directly from structured enterprise data. This role focuses on building machine learning systems for structured and tabular data rather than general LLM application development. The Applied AI Research Team sits at the intersection of theory and production. Your work will take ideas emerging from fundamental research and turn them into practical algorithms, optimized pipelines, and production-ready ML systems that operate across petabytes of structured enterprise data. This is a high-ownership role for engineers who can think like researchers and build like systems engineers. You will translate theory into measurable performance improvements and help define the engineering foundations of structured AI.

Requirements

Strong background in machine learning, probabilistic modeling, optimization, or large-scale ML systems
Experience building algorithms for structured, relational, tabular, or graph data
Ability to reason from first principles about scaling behavior, efficiency, and information flow
Hands-on experience with PyTorch, JAX, TensorFlow, or similar ML frameworks
Strong programming skills in Python
Proven ability to turn research ideas into performant, reliable code
Comfort working in research-driven environments with ambiguous problem definitions
Strong experimentation discipline and focus on measurable performance improvements

Nice To Haves

Experience with systems languages such as Rust, C++, or CUDA is a plus
Experience building large-scale ML pipelines, evaluation frameworks, or distributed systems
Structured representation learning, tabular ML, relational learning, or graph ML
Experience with large-scale training infrastructure or distributed ML
Familiarity with data systems, query engines, or large-scale data pipelines
Experience building evaluation infrastructure for ML systems
Open-source contributions or collaborative work bridging research and production systems

Responsibilities

Turn research into working systems
Transform foundational ideas from Granica Research and Prof. Andrea Montanari’s group into scalable algorithms and prototypes
Build evaluation harnesses, datasets, and benchmarks that measure real signal from research ideas
Define and improve metrics that quantify progress in structured AI systems
Invent and optimize algorithms
Develop efficient learning methods for relational, tabular, graph, and enterprise datasets
Prototype representation learning architectures and compression-aware models
Explore new approaches for learning from heterogeneous structured data
Build high-performance ML pipelines
Implement fast training and inference pipelines using PyTorch, JAX, or custom kernels
Optimize memory usage, compute utilization, and data movement
Improve cost, latency, and throughput for large-scale ML workloads
Build hybrid AI systems
Design systems integrating symbolic, relational, and neural components
Enable AI models to reason over structured datasets without relying on text intermediaries
Collaborate across research and engineering
Work with Research Scientists to validate hypotheses at scale
Work with Systems Engineers to integrate algorithms into Granica’s data platform
Work with Product Engineering to ship features powering real enterprise workloads
Iterate fast and measure everything
Run controlled experiments and analyze performance improvements
Deliver results with clear benchmarks and reproducible evaluations
Drive the cycle from prototype → production → optimization