AI Toolchain Algorithm Engineer (NPU)

Hellosg Recruitment; Singapore; Hybrid; SGD 78,000 - 134,000; Full time

Job summary

Hellosg Recruitment is seeking an experienced AI Toolchain Algorithm Engineer to develop and optimize AI toolchains for on-chip NPUs in Singapore. You will work on AI operator development, acceleration, and performance optimization to enable efficient AI workloads on NPU-based hardware.

The role requires a Master’s degree and 3+ years in AI toolchain or compiler development, with hands-on NPU experience. English and Chinese bilingual communication is essential for collaboration with the

Qualifications

• Master's degree or higher in a relevant field (CS/EE/Automation/Math/Physics).
• 3+ years in AI toolchain or compiler development or related fields.
• Hands-on experience with NPU toolchains and hardware-specific optimization.
• Strong deployment and optimization for AI models on specialized hardware.
• Experience in AI compiler development or operator optimization preferred.
• Strong problem-solving and teamwork in a fast-paced environment.
• Bilingual English and Chinese for China-based collaboration.

Responsibilities

• Develop, maintain, and optimize AI toolchains for chip-based NPU platforms.
• Develop, implement, and optimize AI operators/kernels for performance.
• Adapt AI models to NPU hardware and optimize performance and resources.
• Analyze bottlenecks to improve execution efficiency of AI workloads.
• Collaborate with hardware, compiler, and algorithm teams for efficient solutions.

Skills

AI toolchain

Compiler engineering

Operator/kernel dev

NPU optimization

Model deployment

Analytical skills

Communication

Bilingual EN/CN

Education

Master's degree

Tools

NPU toolchains

AI optimization tools

Job description

AI Toolchain Algorithm Engineer (NPU)

Location: Singapore (Flexible - Work From Home or Office)

Salary Range: SGD 7,000 - 12,000 per month (Negotiable)

Job Summary

We are looking for an experienced AI Toolchain Algorithm Engineer to develop and optimize AI toolchains for on-chip Neural Processing Units (NPUs).

The successful candidate will work on AI operator development and acceleration, model adaptation, and performance optimization to enable efficient AI workloads on NPU-based hardware.

Key Responsibilities

• Develop, maintain, and optimize AI toolchains for chip-based NPU platforms.

• Develop, implement, and optimize AI operators/kernels to improve computational efficiency and hardware performance.

• Adapt AI models to NPU hardware and optimize model performance, computational efficiency, and resource utilization.

• Analyze computational bottlenecks and improve the execution efficiency of AI workloads on NPU platforms.

• Work closely with hardware, compiler, and algorithm teams to deliver efficient AI computing solutions.

Requirements

• Master's degree or above in Computer Science, Electrical/Electronic Engineering, Automation, Mathematics, Physics, or a related discipline.

• 3+ years of relevant experience in AI toolchain development, AI compiler engineering, AI operator/kernel development, or related fields.

• Hands-on experience with NPU toolchains and a strong understanding of AI algorithms and hardware-specific operator optimization.

• Solid knowledge of AI model deployment and optimization on specialized AI hardware.

• Experience in AI compiler development, operator/kernel optimization, or related research and projects is highly preferred.

• Strong analytical, problem-solving, communication, and teamwork skills, with the ability to work effectively in a challenging and fast-paced environment.

• Bilingual proficiency in English and Chinese is required to communicate and collaborate effectively with the China-based team.