AI Infrastructure Engineer

Overview

In this role you will help build and scale Intercom’s AI infrastructure, enabling fast training and reliable inference for large transformer and LLM models. You will work with a small, highly technical team to push the performance and reliability of our AI platform, powering Fin and the Intercom Customer Service Suite. You’ll collaborate with ML scientists to bring cutting-edge training and inference methods into production and mentor other engineers. This is a hands-on, impact-driven role at the core of our AI strategy.

Pay / Benefits

• Competitive salary and equity
• Lunch and snacks
• Performance reviews
• Access to Claude Code and AI tools
• Pension scheme with 4% match
• Health and dental insurance for you and dependents

Responsibilities

• Implement and scale training pipelines for large transformer and LLM models from data ingestion to distributed training and evaluation
• Build and optimize low-latency, reliable inference services with autoscaling, routing, and fallbacks
• Tune GPU kernels, improve utilization, and identify bottlenecks across training and inference stacks
• Collaborate with ML scientists to productionize advanced training and inference methods
• Mentor and develop other engineers on the team and elevate technical standards and reliability across the AI platform

Key requirements

• 5+ years of software engineering with a track record of shipping high-quality products or platforms
• Degree in Computer Science, Computer Engineering, or related field (or equivalent experience)
• Hands-on experience with model training (transformers/LLMs) or model inference at scale
• Experience with low-level GPU work (CUDA, Triton)
• Production experience at meaningful scale and strong communication skills
• Proficiency with at least one programming language (Python, Ruby, Java, Go, etc.)

• Clear communication
• Collaborative mindset
• Continuous learner
• Model training with transformers/LLMs
• Model inference at scale
• GPU programming (CUDA, Triton)

…