Olix

3 months ago

Architect/Staff Systems Software Engineer

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI

Last checked: 14 hours agoStatus: Live

Job Summary

Own the DX-1 accelerator production inference platform by building and extending the distributed runtime and serving stack that connects PyTorch/JAX to the hardware. Drive how inference scales across rack-scale disaggregated topology, with focus on parallelism, KV-cache management, scheduling, reliability, and observability. Lead pre-silicon bring-up using simulation/emulation/FPGA and shape standards and technical direction across cross-functional teams.

Location: London

Workplace: Onsite

Employment Type: Full time

Job Function: Software Engineering

Seniority: Sr. Manager level

Key Responsibilities

•Design, build, and extend the distributed inference and serving stack onto DX-1 without treating layers as black boxes.
•Define how inference scales across many accelerators, covering parallelism strategies, collective communication, KV-cache/offload, and memory-aware scheduling.
•Engineer reliability at scale across failure domains, including fault handling, graceful degradation, load balancing, recovery, and observability standards.
•Evaluate system behavior during bring-up (simulation, emulation, FPGA prototyping, analytical modeling), root-cause failures, and influence hardware/software design decisions.
•Set technical standards across teams through reviews/pairing, and bring structure and direction to ambiguous cross-team platform problems.

Pay and Benefits

Salary: GBP 387,000 annually

Equity and Bonus:Equity

Perks:EquityPension

Key Requirements

•Hands-on systems software experience with strong C/C++ and fundamentals across the runtime/network/accelerator boundary.
•Proven end-to-end ownership of a hard systems problem, ideally extending a distributed inference/serving stack (e.g., vLLM, SGLang, NVIDIA Dynamo, TensorRT-LLM) in production.
•Experience scaling distributed inference, including tensor/pipeline/data parallelism, collective communication, KV-cache and memory management, and reliability across failure domains.
•Fluency at the framework boundary, connecting accelerators to PyTorch/JAX and serving stacks without treating them as opaque.
•Bachelor’s degree or higher in computer science, electrical engineering, mathematics, or a related field.

Experience:AI infrastructureDistributed systemsMachine learningInferenceHPC

Education:Bachelor's

Skills:OwnershipBusiness-aware judgmentStructured problem-solvingCommunicationCross-functional collaboration

Tech Stack:CC++VLLMSGLangNVIDIA DynamoTensorRT-LLMPyTorchJAXFPGAASICPrometheusGrafanaSRAM

Company Brief

Olix

Olix designs and manufactures optical processors (Optical Tensor Processing Units) for training and running inference on large AI models, aiming to deliver higher performance and lower energy consumption than conventional silicon-based accelerators.

Industry: Deep Tech

Company Size: Medium (51 to 250 employees)

Growth: Early Stage Startup

Headquarters: London, United Kingdom

Founded: 2024

Glassdoor: 3.0

Website LinkedIn Glassdoor

Loading job details...

Preparing the role view and application actions.

Olix

3 months ago

Architect/Staff Systems Software Engineer

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI

Last checked: 14 hours agoStatus: Live

Job Summary

Location: London

Workplace: Onsite

Employment Type: Full time

Job Function: Software Engineering

Seniority: Sr. Manager level

Key Responsibilities

•Design, build, and extend the distributed inference and serving stack onto DX-1 without treating layers as black boxes.
•Define how inference scales across many accelerators, covering parallelism strategies, collective communication, KV-cache/offload, and memory-aware scheduling.
•Engineer reliability at scale across failure domains, including fault handling, graceful degradation, load balancing, recovery, and observability standards.
•Evaluate system behavior during bring-up (simulation, emulation, FPGA prototyping, analytical modeling), root-cause failures, and influence hardware/software design decisions.
•Set technical standards across teams through reviews/pairing, and bring structure and direction to ambiguous cross-team platform problems.

Pay and Benefits

Salary: GBP 387,000 annually

Equity and Bonus:Equity

Perks:EquityPension

Key Requirements

•Hands-on systems software experience with strong C/C++ and fundamentals across the runtime/network/accelerator boundary.
•Proven end-to-end ownership of a hard systems problem, ideally extending a distributed inference/serving stack (e.g., vLLM, SGLang, NVIDIA Dynamo, TensorRT-LLM) in production.
•Experience scaling distributed inference, including tensor/pipeline/data parallelism, collective communication, KV-cache and memory management, and reliability across failure domains.
•Fluency at the framework boundary, connecting accelerators to PyTorch/JAX and serving stacks without treating them as opaque.
•Bachelor’s degree or higher in computer science, electrical engineering, mathematics, or a related field.

Experience:AI infrastructureDistributed systemsMachine learningInferenceHPC

Education:Bachelor's

Skills:OwnershipBusiness-aware judgmentStructured problem-solvingCommunicationCross-functional collaboration

Tech Stack:CC++VLLMSGLangNVIDIA DynamoTensorRT-LLMPyTorchJAXFPGAASICPrometheusGrafanaSRAM

Company Brief

Olix

Industry: Deep Tech

Company Size: Medium (51 to 250 employees)

Growth: Early Stage Startup

Headquarters: London, United Kingdom

Founded: 2024

Glassdoor: 3.0

Website LinkedIn Glassdoor