Inference Engineer, AGI

Description

We are looking for an Inference Engineer to help serve inference for real-time multimodal conversational AI. This is a full-stack inference role: you will work across the path a model takes from research to production, helping shape model architecture so it is servable, building the real-time runtime that serves it within hard latency budgets, and building the offline systems that train and reinforce it.

You will operate at the boundary of Science and Inference, taking frontier-scale speech and audio models and helping make them run within real-time latency budgets on production hardware. You will help co-design architectures with scientists to make them inference-friendly, contribute to the low-latency streaming serving path, and help build the training and reinforcement-learning infrastructure that closes the loop. You will have the compute, data, and runway to work on problems that few teams in the world are positioned to tackle.

As an Inference Engineer, you will own well-scoped components of the inference stack, drive their technical execution with guidance, and work closely with scientists and senior engineers to help ensure our models run fast enough to feel human in real time, and at a cost that makes them viable  may go deep in one of the areas below while contributing across the others.

Key job responsibilities

Model Architecture & Inference Co-Design

- Partner with research scientists and senior engineers to help make model architectures servable, surfacing the latency, memory, and cost implications of architecture choices

- Implement and optimize parts of the inference path for large-scale multimodal models, attention and KV-cache mechanisms, multimodal/autoregressive decoding, and the compute primitives on the critical path

- Apply efficiency techniques across the stack, quantization (per-tensor/per-channel/per-group, INT8/FP8/BF16), speculative decoding, operator fusion, and paged KV-cache, and measure their quality/latency trade-offs

- Help develop and tune high-performance kernels for critical operations where off-the-shelf implementations leave performance on the table, integrating them into production serving

- Profile end-to-end performance with tools such as Nsight Compute/Systems and roofline analysis to help identify and eliminate bottlenecks in large-scale inference workloads

Real-Time & Interactive Runtime

- Contribute to the real-time serving path for streaming multimodal conversational AI, helping meet sub-second, streaming latency budgets under concurrent session load

- Build and tune continuous batching, scheduling, and preemption to balance throughput against per-request latency SLAs for interactive workloads

- Customize production serving frameworks (e.g., vLLM, PyTorch) for real-time streaming generative models that fall outside standard LLM serving patterns

- Implement multi-GPU inference (tensor parallelism, collective communication) for latency-critical paths, and help drive cost toward parity with existing production baselines

- Help establish latency, throughput, and cost benchmarking, and publish the operational metrics that gate deployment

Offline Systems:
Training, RL & Evaluation Infrastructure

- Build and scale parts of the offline inference systems behind post-training, high-throughput rollout generation and reward-model serving for reinforcement learning (RL/RLHF/RLAIF)

- Help ensure train/serve consistency, that the inference path used in RL and evaluation faithfully matches production online behavior (e.g., parity across sampling and logit processing)

- Work with the evaluation team to enable offline inference that captures the quality dimensions unique to real-time conversation, latency sensitivity, audio quality, and interaction naturalness

Basic Qualifications

- Master's degree or equivalent

- 3+ years of non-internship professional software development experience

- 3+ years of programming with at least one software programming language experience

- 2+ years of experience contributing to the design or architecture (design patterns, reliability and scaling) of new and existing systems

- Bachelor's degree in computer science or equivalent

- 1+ years of hands-on experience optimizing inference for neural models, not just using inference frameworks, but profiling and improving them

- Solid understanding of deep learning architectures (transformers, attention mechanisms, autoregressive decoding) and their application to speech/audio or other multimodal domains

- Experience contributing to…