AI Inference Engineer

About the role

Paranine is seeking an AI Inference Engineer to design, implement, and optimise high‑performance inference pipelines for large language models on Huawei NPUs and Nvidia GPUs. The role focuses on profiling, tuning, and maintaining production‑grade services that deliver low latency, high throughput, and cost‑effective AI inference.

Key responsibilities

• Develop and maintain orchestration services for model deployment and inference.
• Profile and optimise resource utilisation on GPU/NPU hardware.
• Implement monitoring, logging, and cost‑attribution tooling across the inference stack.
• Troubleshoot production issues and ensure nines‑level uptime for regional deployments.

Required profile

• Strong foundation in computer science concepts, algorithms, data structures and systems programming.
• Hands‑on experience building production‑grade services and APIs for distributed systems.
• Expertise in neural networks, pattern recognition and optimisation of inference workloads.
• Practical knowledge of NLP and scaling large language models.

Required skills

• Programming languages: Go, Python, C++, Rust.
• GPU/NPU acceleration frameworks: CUDA, ROCm, vendor SDKs.
• Containerisation and orchestration: Kubernetes, Docker.
• Monitoring and observability: metrics, logging, tracing, performance profiling.