
Principal AI Engineer
Posted 1 day ago

Posted 1 day ago
This is a fully remote position, open to applicants in Ireland.
• Outline the architecture and technical roadmap for LLM and agentic AI systems within Detection and Response, as well as Exposure and Remediation.
• Oversee model strategy, determining when to utilize frontier or open-source models and the appropriate times to fine-tune, distill, or post-train.
• Fine-tune, distill, and post-train open-source LLMs using security data, guiding the process from data curation through to training, evaluation, and deployment.
• Establish engineering standards for context engineering, inference optimization, and evaluation across various teams.
• Enhance LLM inference at scale through methods such as speculative decoding, KV-cache management, continuous batching, quantization, and custom kernels.
• Implement evaluations as the essential release gate for every AI system deployed by Rapid7.
• Facilitate design reviews and address the most challenging technical trade-offs.
• Prototype and develop essential model, training, and inference code in a hands-on capacity.
• Provide mentorship to Staff and Senior engineers.
• Influence product and platform investment decisions.
• Shape how Rapid7 develops and scales AI throughout the organization.
• A minimum of 12 years of experience in software or ML engineering, including at least 2 years focused on building production LLM or agentic AI systems.
• A preferred PhD in Computer Science, Machine Learning, AI, or a related field; alternatively, equivalent expertise demonstrated through industry experience and technical contributions.
• Proven record of developing scalable, production-grade AI systems utilized across various teams or products.
• A strong history of making organization-wide impacts through technical leadership, mentorship, and establishing shared platforms and standards.
• In-depth knowledge of transformer architecture, attention mechanisms, tokenization, positional encodings, as well as model training and serving; capable of implementing core components from the ground up.
• Expertise in PyTorch, including custom model code, training loops, and mixed precision and distributed training techniques like FSDP and DeepSpeed.
• Practical experience with full and parameter-efficient fine-tuning, LoRA, QLoRA, distillation, and preference or RL-based techniques such as DPO and GRPO.
• Skills in context engineering that involves retrieval, memory, and context management for sustained agent workflows.
• Profound knowledge of prefill and decode processes, KV-cache management, PagedAttention, prefix caching, continuous batching, speculative decoding, FlashAttention, and serving at scale with vLLM, SGLang, or TensorRT-LLM.
• Experience in defining evaluation strategies and benchmarks for LLM and agent systems.
• Strong foundational knowledge in predictive AI and classical ML.
• Proficiency in distributed systems and Python fundamentals.
• Capability to take ownership of significant technical outcomes across multiple teams.
• Excellent communication skills for articulating technical trade-offs to engineering, product, and leadership audiences.
• Ability to forge cross-functional partnerships across Engineering, Product, and other teams.
• Competence in aligning customer needs and business priorities with technical decisions.
• Ability to embody the core values of Rapid7.
• Preferred: experience in pre-training or training LLMs from the ground up.
• Preferred: experience in writing GPU kernels using CUDA or Triton, along with profiling GPU workloads.
• Preferred: background in the security domain such as detection engineering, SOC/MDR, or vulnerability management.
• Preferred: experience in running AI workloads on AWS at scale.
• Preferred: publications, patents, or open-source contributions in the field of AI.
• Variable/incentive compensation (amount not specified).
• Equity (amount not specified).
• Benefits (details not specified).
• Fair and equitable compensation practices.
Extend
OhioHealth
Capital One
Software Mind
Get handpicked remote jobs straight to your inbox weekly.