AI Performance Engineer

Job not on LinkedIn

πŸ•’ October 30

πŸ‡¬πŸ‡§ United Kingdom – Remote

πŸ’΅ Β£99.6k - Β£150k / year

⏰ Full Time

🟑 Mid-level

🟠 Senior

πŸ€– Artificial Intelligence

Where this job comes from

Posted on Modular's careers page β†—Modular's careers page, not through a job board

First seen by Remote Rocketship October 30

Still open on Modular's site. Checked 4 hours ago.

Apply Now
Logo of Modular

Modular

51 - 200 employees

πŸ€– Artificial Intelligence

🏒 Enterprise

πŸ’° $100.5G Series B on 2023-08

Artificial Intelligence β€’ Enterprise

Modular is the next-generation AI developer platform unifying the development and deployment of AI for the world. The company focuses on enabling AI to be accessible to anyone, anywhere, leveraging specialties in machine learning, AI technologies, and platforms such as TensorFlow and PyTorch, along with hardware solutions.

πŸ“‹ Description

β€’ Design and implement high-performance attention kernels for prefill and decode, including MHA, GQA, paged attention, sliding-window/global attention, and fused FlashAttention-style algorithms β€’ Build and optimize general AI kernels, including matrix multiplication, softmax, normalization, activation, embedding, reduction, and data-movement primitives β€’ Contribute to the design and implementation of Modular’s Attention framework β€’ Profile workloads, identify bottlenecks, establish performance models, and tune implementations against hardware limits and competitive baselines β€’ Develop benchmarks, correctness tests, regression tests, and automated performance analysis β€’ Partner with Modular teams on software stack improvements and share learnings β€’ Work with hardware partners to understand architectures, validate toolchains, diagnose low-level issues, and influence hardware/software interfaces β€’ Write design documents and participate in technical reviews β€’ Share architecture and performance insights across the team β€’ Mentor teammates and improve kernel quality, performance methodology, and maintainable low-level software β€’ Independently own ambiguous performance problems and turn research ideas and hardware capabilities into robust product technology

🎯 Requirements

β€’ 5+ years of relevant industry or research experience in high-performance computing, AI kernel development, accelerator programming, compiler engineering, or a closely related field β€’ Demonstrated experience writing and optimizing production-quality GPU or accelerator kernels β€’ Strong understanding of parallel algorithms, numerical behavior, memory access patterns, synchronization, and data movement β€’ Strong knowledge of modern attention algorithms and implementation tradeoffs, including prefill and decode performance characteristics β€’ Strong understanding of architectures beyond conventional GPUs, such as NPUs, DSPs, TPUs, or custom ASICs β€’ Experience with at least one heterogeneous programming model or kernel ecosystem such as CUDA, Triton, SYCL, OpenCL, HIP, or a vendor accelerator SDK β€’ Strong performance-analysis skills and proficiency with profiling, tracing, benchmarking, and debugging tools β€’ Ability to read architecture manuals, compiler output, and low-level generated code β€’ Ability to form hypotheses from data and systematically close performance gaps β€’ Collaborative, team-oriented approach; clear written and verbal communication; intellectual curiosity; and comfort owning ambiguous technical problems β€’ Knowledge of Mojo programming language is helpful but not required β€’ Knowledge of MLIR, LLVM, and related AI compiler technologies is helpful but not required β€’ Experience implementing FlashAttention or other fused attention algorithms is helpful but not required β€’ Experience with modern kernel DSLs and libraries such as CuTe/CUTLASS, Triton, Pallas, or similar systems is helpful but not required β€’ Familiarity with AI framework internals such as PyTorch, JAX, TensorFlow, vLLM, SGLang, or TensorRT-LLM is helpful but not required β€’ Experience bringing up a model or kernel library on a new hardware platform is helpful but not required β€’ Experience with performance modeling, autotuning, numerical validation, or benchmarking infrastructure is helpful but not required β€’ Familiarity with LLM architectures such as GPT and Gemma is helpful but not required β€’ Advanced degree in Computer Science, Electrical Engineering, Computer Engineering, or a related field is helpful but not required

πŸ–οΈ Benefits

β€’ Comprehensive healthcare coverage β€’ Retirement and savings programs β€’ Employee stock purchase opportunities β€’ Paid time off β€’ Wellbeing resources β€’ Family support programs β€’ Learning and development opportunities β€’ RSU grants β€’ Annual target bonus β€’ Equity β€’ Regular team onsites and local meetups β€’ Travel 2–4 times a year β€’ Remote work from home option β€’ In-person onboarding at the appropriate office

Apply Now

πŸ“Š Check your resume score for this job

Improve your chances of getting an interview by checking your resume score before you apply.

Similar Jobs

πŸ•’ October 29

UFORCE

501 - 1000

πŸŽ–οΈ Defense

πŸ”§ Hardware

πŸš€ Aerospace

AI Specialist building AI workflows for Marketing, Military Exercises, Sales, and RevOps at UFORCE, a London-based defence-technology company. Improving GTM capacity, quality, speed, and information access.

πŸ‡¬πŸ‡§ United Kingdom – Remote

πŸ”₯ Funding within the last year

πŸ’° $50M Seed Round on 2026-03

⏰ Full Time

🟑 Mid-level

🟠 Senior

πŸ€– Artificial Intelligence

πŸ•’ October 29

ServiceNow

10,000+ employees

🏒 Enterprise

☁️ SaaS

πŸ€– Artificial Intelligence

Staff AI Agent Engineer deploying Moveworks ServiceNow AI solutions for enterprise customers. Designing integrations, automations, and agentic workflows while shaping product evolution through customer feedback.

πŸ•’ October 29

Paddle

201 - 500

☁️ SaaS

πŸ’³ Fintech

🀝 B2B

Organic and AI Search Manager improving discoverability for Paddle, a Merchant of Record payment infrastructure company. Driving SEO, LLM citations, content strategy, and commercial pipeline impact.

πŸ•’ October 29

Storm Ideas

51 - 200

πŸ“£ Marketing

πŸ’Ό Consulting

πŸ₯ Healthcare

Creative Operations Officer building Airtable, automation and AI-agent workflows for Storm Ideas. Improving creative operations for globally recognised US TV and entertainment brands.

πŸ•’ October 29

Think Jam

11 - 50

πŸ’Ό Consulting

πŸ“£ Marketing

πŸ“± Media

VP of AI transforming Think Jam’s entertainment creative agency through multi-agent systems. Designing scalable workflows, integrations, and guardrails for social, creative, analytics, and video products.

Developed byΒ Lior Neu-ner. I'd love to hear your feedback β€” Get in touch via DM or support@remoterocketship.com
Follow Remote Rocketship on LinkedInFollow Remote Rocketship on InstagramFollow Remote Rocketship on FacebookFollow Remote Rocketship on TikTok