Member of Technical Staff (Research Infrastructure)

Location: San Francisco, United States; on-site

Salary: $200,000–$400,000 USD per year, plus equity and comprehensive benefits.

Company

hirify.global builds AI infrastructure for simulating human behavior at scale and helping organizations make decisions using verifiable predictions.

What you will do

• Build and operate the ML platform covering data exploration, feature generation, experiment tracking, training orchestration, evaluation, and deployment.
• Optimize data and training pipelines by profiling FLOPs utilization, tokenization, GPU memory, ingestion throughput, and system observability.
• Improve population-scale inference through batching, scheduling, KV-cache reuse, quantization, and other serving optimizations.
• Redesign simulation data architecture, including schemas and ingestion logic for high-variety behavioral data.
• Own the multi-node GPU cluster, including NCCL, RDMA, storage, scheduling, autoscaling, portability, and alerting.
• Build rigorous statistical evaluation tooling and translate research advances in simulation, training, and inference into production improvements.

Requirements

• Deep systems and ML proficiency, including high proficiency in Python and hands-on experience with PyTorch or JAX.
• Experience optimizing modern ML architectures and working with NVIDIA GPUs and technologies such as NCCL, InfiniBand, NVLink, CUDA, or Triton.
• Experience building production ML platforms or MLOps systems for training orchestration, experiment tooling, model serving, or LLM applications.
• Experience architecting, observing, debugging, and scaling production distributed systems, ideally performance-critical systems.
• Understanding of the training, fine-tuning, continuous ingestion, monitoring, and high-availability serving lifecycle.
• Strong quantitative foundation and research/data literacy; a background in computer science, mathematics, statistics, or a related field is typical.

Nice to have

• Experience optimizing inference for multi-agent or agentic environments with interdependent requests.
• Experience with distributed training at scale.
• Experience shipping production AI agents and familiarity with vLLM, SGLang, TensorRT-LLM, or inference-time optimization.
• Interdisciplinary experience in social science modeling or behavioral economics.
• Experience writing custom GPU kernels.

Culture & Benefits

• Competitive compensation combining base salary, equity, and comprehensive benefits.
• Medical, dental, and vision coverage.
• Flexible time off supporting work-life balance.
• Emphasis on thoughtful conversations, clear examples of past work, humility, collaboration, and pragmatic ownership.

Hiring process

• The hiring journey uses thoughtful conversations and clear examples of past work to assess mutual fit, working style, and expectations.
• Candidates may reapply for the same role after a 90-day waiting period.