Research Engineer – Inference

This is a fully remote position, open to applicants in United Kingdom.

📋 Description

• Implement cutting-edge AI models into production environments

• Take ownership of the transition from research milestones to serving infrastructure

• Enhance inference performance focusing on latency, throughput, and cost efficiency

• Utilize quantization, distillation, KV-cache optimization, batching strategies, and custom kernels

• Develop and optimize high-performance serving systems for real-time and streaming workloads

• Establish tools and infrastructure that enable researchers to deploy models to production swiftly and securely

• Assess and validate the performance characteristics of models

⛳️ Requirements

• No formal qualifications or degrees necessary

• Passion for tackling challenging engineering problems

• Capability to showcase work through previous projects, designs, or contributions on GitHub

• Experience in deploying and serving ML models in production, particularly for latency-sensitive or real-time applications

• Strong engineering expertise in GPU programming and inference optimization, including CUDA, Triton, TensorRT, vLLM, or SGLang

• Ability to independently profile, diagnose, and resolve bottlenecks throughout the serving stack

• Skill in developing tools to measure the performance of the serving stack

🏝️ Benefits

• Annual discretionary professional development stipend

• Annual discretionary stipend for social travel to connect with colleagues

• Annual company offsite

• Monthly co-working stipend for employees residing away from a main hub

• Option to work from company offices located in London, New York, San Francisco, and Warsaw