Lead AI/LLM Inference Engineer (Go/Python) for Production

JPMorgan Chase & Co.; Glasgow; On-site; GBP 90,000 - 130,000; Full time

Job summary

JPMorganChase in Glasgow is seeking a Lead Software Engineer to advance LLM inference performance within the AI/ML Data Platform.

You will benchmark, tune, and optimize production workloads, collaborating with senior engineers to deliver fast, cost-efficient, production-ready models across the firm's AI systems.

This role blends cutting-edge ML with large-scale systems, offering impact across the enterprise while growing expertise in quantization, scheduling, and GPU optimization.

Qualifications

• Formal training or certification in software engineering concepts; preferably Go / Python.
• Hands-on experience with LLM inference systems in production environments.
• Strong understanding of GPU memory architecture and quantization implications at inference time.

Responsibilities

• Execute benchmarking and performance characterization across production LLM workloads; establish baselines and measure impact of changes.
• Design and run quantization experiments (FP8, INT8/INT4) to evaluate accuracy, throughput, and cost-per-token.
• Support speculative decoding strategy; provide configuration recommendations for multiple workloads.
• Build and maintain GPU efficiency metrics and provide a data-driven view of platform efficiency.
• Benchmark the platform against external providers and industry numbers; identify gaps and drive improvements.
• Participate in inference engine upgrade evaluations and validation before production promotion.
• Contribute to GPU chaos testing and hardware diagnostics; measure recovery and resilience.
• Leverage enterprise AI-assisted development tools to improve code quality and productivity; validate AI outputs.

Skills

Go/Python

LLM inference systems

GPU memory architecture

Quantization techniques

Benchmarking

Cloud GPU infra

Tools

GuideLLM

TensorRT-LLM

Job description

JPMorganChase in Glasgow is seeking a Lead Software Engineer to advance LLM inference performance within the AI/ML Data Platform.

You will benchmark, tune, and optimize production workloads, collaborating with senior engineers to deliver fast, cost-efficient, production-ready models across the firm's AI systems.

This role blends cutting-edge ML with large-scale systems, offering impact across the enterprise while growing expertise in quantization, scheduling, and GPU optimization.