Software Engineer, Machine Learning Platform - Gen AI
Core
Building production infrastructure for Generative AI, specifically open-weight LLMs and VLMs, enabling real-time GPU serving, high-throughput batch inference, and model fine-tuning for DoorDash, Wolt, and Deliveroo.
Role type
Senior Machine Learning Platform Engineer (Gen AI Infrastructure)
Builds
Open-weights model serving stack, inference engines, fine-tuning pipelines, GPU autoscaling, and platform primitives for GenAI products.
Domain
Generative AI, Large Language Models (LLMs), GPU Infrastructure, Distributed Systems
Deliverable
production ML models
Required skills
Python, distributed systems, LLM inference, model fine-tuning (SFT/DPO/LoRA), GPU utilization optimization, observability, incident response, AI coding tools.
Preferred skills
LLM inference engines (vLLM, SGLang, TensorRT-LLM), distributed fine-tuning, GPU performance tuning (quantization, KV-cache), Kubernetes, cloud infrastructure (AWS/GCP), AI agents, eval systems.
Technologies
Python, vLLM, SGLang, TensorRT-LLM, Kubernetes, AWS, GCP, CUDA, FP8, INT8, AWQ, GPTQ
Responsibilities
Design scalable systems for model serving and batch inference; optimize cost and latency of GPU inference; build platforms supporting rapid experimentation; partner with ML engineers and data scientists to turn capabilities into platform primitives; shape future directions like RLHF and agent optimization.
Seniority
Senior, hands-on IC
