AI Engineer

Description

The team builds the core systems behind model training, deployment and inference at scale, supporting AI across the company's products, including real-time communication, computer vision and natural language understanding.

This role centres on the inference systems that serve the company's AI models in production. You will work with senior inference engineers and with research and platform teams.

Your work is to measure serving workloads, locate what holds back latency, throughput and cost, and deliver the changes that fix it, working deep inside inference frameworks, GPU behaviour and serving infrastructure.

The company makes collaboration software that helps enterprise customers communicate across contact center, phone, events, apps, rooms and webinars.

The team moves quickly, designs with customers and users in mind, and offers room to stretch your skills and grow your career.

Responsibilities

• Build, tune and maintain parts of the inference serving stack for transformer and multimodal models, including large and Mixture-of-Experts (MoE) designs.
• Implement and assess inference optimisations such as speculative decoding, continuous batching, KV cache management, parallelism layouts and INT4, INT8 or FP8 quantisation.
• Extend and adapt inference frameworks such as vLLM, SGLang, TensorRT-LLM or equivalents to support new models and the company's production needs.
• Profile GPU workloads with Nsight Systems or similar tools, trace the root cause of performance regressions, and confirm gains with repeatable benchmarks.
• Create and improve internal tools for benchmarking, performance modelling and serving configuration.
• Work with researchers to move new model architectures into production efficiently.
• Share in on-call and production support for the services you build, with guidance from senior engineers.

Requirements

• A degree in Computer Science, Computer Engineering or a related field, or equivalent hands-on experience. An MS or PhD is preferred.
• 0, 3 years of software engineering experience. Internships, research and open-source work on ML systems count.
• Hands-on work with at least one inference framework (vLLM, SGLang, TensorRT-LLM or ONNX Runtime), including serving and modifying it, not just exporting models.
• Working understanding of LLM inference on GPUs, covering memory bandwidth, KV cache, batching and communication across GPUs.
• Experience profiling and debugging GPU performance, with a measured optimisation you have shipped.
• Depth in at least one of: speculative decoding, continuous batching, KV cache design, quantisation, or distributed and multi-GPU serving.
• Strong Python and C++ skills, and comfort reading and changing large framework codebases.
• Preferred: merged contributions to vLLM, SGLang, TensorRT-LLM, PyTorch or similar projects.
• Preferred: CUDA, Triton or CUTLASS kernel development.
• Preferred: experience serving MoE or 100B+ parameter models, or multi-node inference using NCCL or RDMA/RoCE.
• Preferred: graduate coursework or research in ML systems, compilers or high-performance computing.

Conditions

Base salary range: $137,700.00 - $275,400.00.

Total direct compensation considers base salary, bonus and equity value, in addition to the base salary and/or OTE listed.

Starting pay depends on several factors and reflects your qualifications and experience.

Pay ranges vary by location, so candidates in other locations may see a different range.

Applications stay open for at least 5 days. The anticipated close date is 11/13/26.

Interviews are run with a consistent interview tool and may be recorded.

The company offers equal opportunity in hiring, judging every candidate on skills, experience and potential.

Benefits

• Benefits cover a range of perks and options that support physical, mental, emotional and financial health, work-life balance, and community involvement.