Edge AI Engineer: SLM & VLM for Low-Latency Inference

Jabil; Singapore; On-site; SGD 180,000 - 240,000; Full time

Job summary

Jabil in Singapore seeks a highly capable engineer/researcher to lead the R&D of Small Language Models (SLMs) and Vision-Language Models (VLMs) for edge, low-latency production scenarios. You will own continuous pretraining, instruction tuning (SFT), and compression/distillation pipelines, collaborating with platform teams to deliver reliable gains in inference efficiency and model quality.

Experience with PyTorch, HPC, and tool-call systems is expected; able to communicate in English and

Qualifications

• Strong software eng in Python and C++, with PyTorch training/eval pipelines
• Hands-on model efficiency and inference optimization experience
• Proficient in HPC concepts (CUDA/SIMD) and performance tuning
• Ability to read/reproduce ideas from recent papers and implement methods
• Communication in English and Mandarin with cross-border teams

Responsibilities

• Lead SLM/VLM training and continuous pretraining workflows
• Design compression, distillation, and edge inference optimizations
• Build and maintain a production-grade tool calling framework
• Apply RL/reward modeling for alignment and reliability
• Automate data collection, cleaning, labeling, and dataset versioning
• Rigorous offline/online evaluation and rapid iteration

Skills

Python

C++

PyTorch

ML pipelines

Model optimization

CUDA

Mandarin

Job description

Jabil in Singapore seeks a highly capable engineer/researcher to lead the R&D of Small Language Models (SLMs) and Vision-Language Models (VLMs) for edge, low-latency production scenarios. You will own continuous pretraining, instruction tuning (SFT), and compression/distillation pipelines, collaborating with platform teams to deliver reliable gains in inference efficiency and model quality.

Experience with PyTorch, HPC, and tool-call systems is expected; able to communicate in English and