Performance Modelling Engineer

ZipRecruiter; London; On-site; GBP 60,000 - 90,000; Full time

Job summary

A leading AI company seeks a Performance Modelling Engineer in London to create analytical models guiding architecture evolution. This role involves developing simulators and analyzing performance trade-offs to influence software and hardware design in a fast-paced environment.

Benefits

Competitive salary and stock options

Comprehensive healthcare insurance

25 days PTO plus bank holidays

Private access to an in-house 3D printer

Qualifications

• 3+ years developing performance or power models.
• Ability to read RTL or schematics.
• Experience with cycle-accurate simulators.

Responsibilities

• Design, build, and maintain functional simulators.
• Develop architectural and cycle-accurate models.
• Conduct workload analysis and identify bottlenecks.

Skills

C++

Python

Data Visualization

Machine Learning Frameworks

Computer Architecture Fundamentals

Education

Bachelor’s degree in EE, CS, Physics, Applied Maths

Tools

gem5

SystemC

Python

C++

Job description

Job Description

Company Overview

Flux Computing designs and manufactures optical processors to train and run inference on large AI models. Join us in London to be part of a highly motivated and skilled team that thrives on delivering impact and innovation at speed.

The Role

We’re searching for a Performance Modelling Engineer to create and own the analytical and simulation models that guide OTPU architecture and software evolution. You will develop functional simulators and high-fidelity, cycle-accurate models of our optical compute system. This role is crucial for exploring design trade-offs and providing insights that influence our software, hardware, and optical roadmaps. It sits at the intersection of hardware architecture, software tooling, and machine-learning workload analysis, ideal for an engineer passionate about data-driven decisions and rapid iteration.

Responsibilities

1. Functional Simulator: Design, build, and maintain a functional simulator of the OPTU subsystem and full pipeline.
2. Performance Simulator: Develop and maintain architectural and cycle-accurate models of OPTU subsystems and pipelines. Identify bottlenecks in throughput, latency, and utilization; propose architectural or scheduling improvements.
3. Workload Analysis & Bottleneck Hunting: Instrument benchmarks (LLMs, diffusion, graph workloads) to collect detailed traces.
4. Design-Space Exploration: Run extensive parameter sweeps with your models to understand trade-offs and guide the software, hardware, and optical teams. Summarize findings into clear, quantitative analyses and recommendations.
5. Tooling & Automation: Develop Python/C++ tools for trace parsing, statistical analysis, and visualization. Integrate models into CI pipelines for performance testing with each RTL commit.

Skills & Experience

• 3+ years developing performance or power models for CPUs, GPUs, ASICs, or accelerators.
• Proficient in C++ and Python; experience with discrete-event or cycle-accurate simulators (e.g., gem5, SystemC, or custom in-house).
• Strong understanding of computer architecture fundamentals: memory systems, interconnects, queuing theory, Amdahl/Gustafson analysis.
• Familiarity with machine-learning workloads and frameworks (PyTorch, TensorFlow, JAX).
• Ability to read RTL or schematics and discuss micro-architectural trade-offs with hardware designers.
• Excellent data visualization and communication skills: capable of turning millions of simulation samples into impactful presentations.
• Bachelor’s degree or higher in EE, CS, Physics, Applied Maths, or related fields; advanced degrees are a plus but not required.
• Personal or open-source projects in simulators, ML kernels, or performance analysis are advantageous.

Compensation & Benefits

• Competitive salary and stock options in a rapidly growing AI company.
• Located in our new 5,000 sq. ft. office in Kings Cross, London, in the AI hub.
• All employees are required to work from our London HQ and live within a 45-minute commute; an additional £24,000/year incentive is offered for those within 20 minutes.
• Comprehensive healthcare insurance.
• 25 days PTO plus bank holidays.
• Private access to our in-house 3D printer.

If you are passionate about advancing AI and excel in a high-energy, fast-paced environment, we want to hear from you. Apply now to join Flux and help shape the future of computing.