Machine Learning Engineer – Embedded AI (Qualcomm) – Shanghai | Visa + Salary Insights

📍 Location: Others

🏷 Type: Full-time

Job Intelligence

📍 More in this location:
Browse AI jobs in Others

🏷 Similar roles:
View similar Full-time jobs

🌐 Explore all jobs:
View all AI job listings

Job Overview

This Machine Learning Engineer role at Qualcomm China focuses on embedded AI systems and on-device machine learning optimization for next-generation mobile and edge computing platforms. The position sits within the engineering group responsible for AI acceleration, performance tuning, and deployment of deep learning models on proprietary Qualcomm hardware. The ideal candidate is a strong systems-level engineer with experience in C/C++, Linux/Android environments, and production-grade ML frameworks such as TensorFlow or ONNX. This role is best suited for engineers who understand both hardware constraints and modern deep learning architectures like CNNs, RNNs, and Transformers, and who can translate models into efficient, power-optimized implementations.

📅 Job Timeline & Status

This is a mid-to-senior level embedded AI engineering role currently in an active hiring phase. Qualcomm roles in Shanghai typically operate on rolling review cycles, meaning strong candidates are evaluated continuously rather than after a fixed deadline. Given the competitive nature of AI systems roles, applicants should apply immediately to maximize visibility before pipeline saturation increases.

🌍 Work Eligibility & Location

This role is strongly tied to onsite embedded systems development due to hardware testing and optimization requirements. While Qualcomm supports global mobility, collaboration with hardware teams and device labs in Shanghai makes relocation highly relevant. International candidates with strong ML systems backgrounds and embedded experience are viable, especially those with experience in heterogeneous compute environments.

💰 Salary Intelligence

Compensation is competitive for China-based semiconductor AI roles, especially given Qualcomm’s position in mobile AI acceleration. Engineers with strong deployment experience in ONNX, TensorFlow Lite, and embedded Linux pipelines may land at the higher end of the range. Total compensation may also include performance bonuses and relocation packages, which significantly increase effective earnings.

📊 Role Breakdown

This role centers on building and optimizing machine learning inference systems for Qualcomm’s AI accelerators. Approximately 40% of the work involves model optimization and compression, including quantization, pruning, and graph transformation to ensure models run efficiently on constrained hardware. Another 25% focuses on embedded system integration, working across Linux and Android stacks to ensure smooth deployment across devices. Around 20% involves performance profiling and debugging, identifying bottlenecks in compute, memory, and power consumption using hardware-level analysis tools. The remaining 15% is dedicated to collaboration with cross-functional teams, including hardware architects and application engineers, to align ML models with silicon-level execution capabilities. Engineers will frequently work with C/C++ for performance-critical modules and use Python for model experimentation and pipeline automation. Strong understanding of CNNs, Transformers, and RNN architectures is essential for adapting models to edge inference constraints.

🧩 Required Skills & Fit

📈 Difficulty & Competitiveness

This is a technically demanding role requiring both deep ML knowledge and embedded systems expertise. The combination significantly raises the hiring bar, as candidates must demonstrate ability to bridge algorithmic modeling and hardware-constrained optimization. Expect strong competition from experienced systems engineers and ML engineers transitioning into edge AI.

🚀 Career Impact

This role provides strong exposure to production-grade AI systems at hardware level, a rare specialization in the industry. Engineers often transition into senior AI infrastructure roles, chip-level ML optimization teams, or principal engineering tracks within semiconductor companies and top AI hardware firms.

📋 Key Responsibilities

You will design and optimize machine learning inference pipelines for Qualcomm AI accelerators. Responsibilities include developing high-performance modules in C/C++, optimizing neural networks for embedded environments, and ensuring compatibility with Android and Linux systems. You will analyze performance bottlenecks in compute, memory, and energy usage, applying hardware-aware optimization strategies. Additional duties include integrating models using TensorFlow or ONNX, refactoring system-level codebases for scalability, and collaborating with cross-functional teams to align ML models with silicon execution constraints. A key focus is ensuring power efficiency, latency reduction, and real-time inference stability across devices.

🎯 Application Strategy

🧠 Application Optimization (Adaptive)

This section is personalized by seniority.

🧠 Fit & Positioning Analysis

Evaluate your match before applying.

📅 Application Signals

Hiring momentum is currently high, with continuous candidate intake typical of Qualcomm engineering roles. Competition is expected to be very strong due to the brand’s global reach and the scarcity of embedded ML specialists. Response cycles are typically fast for shortlisted candidates, with initial feedback often within 1–3 weeks. Since no strict deadline is published, applications may close without notice, and early submission significantly improves visibility in recruiter pipelines.

🔗 Apply for this Job

Apply on Company Site

✅ Job Source & Verification

This listing is based on an official Qualcomm careers posting for an embedded Machine Learning Engineer role in Shanghai. Source credibility is high as it originates from Qualcomm’s verified recruitment platform. Last updated: May 31, 2026, reflecting the most recent review of publicly available job information. Candidates should still confirm final details on the official portal prior to applying, as engineering roles may be updated dynamically based on hiring demand.