Amazon's AGI Web & Knowledge Services group is seeking a passionate, talented, and inventive MLE to lead the development of industry-leading multi-modal search systems.
As part of our cutting-edge multi-modal search team, you will play a pivotal role in reinventing efficient AI solutions for a multi-modal future at scale.
In this role, you will work alongside renowned researchers and engineers to enable our customers to seamlessly interact with unstructured and semi-structured content through advanced capabilities like question answering, contextual search, and multi-turn dialogues.
You will design, develop, and deploy scalable and efficient machine learning models and systems to power these transformative multi-modal AI applications.
The scope of these efforts includes defining public APIs, performance tuning and analysis, crafting and implementing compiler and optimization techniques for neural networks, and other general software engineering work.
Key job responsibilities
- Build high-throughput, cost-effective data pipelines to support feature extraction and indexing for our web-scale Information Retrieval system
- Design and build cost-efficient distributed model training infrastructure
- Develop efficient, state-of-the-art streaming algorithms for processing large datasets (e.g. deduplication, topic clustering)
- Build on and maintain an existing code base as well as new components; maintain production code and contribute to deployment and QA processes
- Serve as technical lead for all stages of the software development cycle, including designing and developing new system architecture and improvements
- Participate in prioritization, estimation, and sprint planning
- Work in an Agile / Scrum environment to deliver high quality software against aggressive schedules
BASIC QUALIFICATIONS
- BS / MS in Computer Science or equivalent experience
- Experience programming with at least one software programming language
- Experience in machine learning, data mining, information retrieval, statistics or natural language processing
- 2+ years of non-internship professional software development experience
PREFERRED QUALIFICATIONS
- Experience working with ML frameworks such as Pytorch, TensorRT, Onnx, AWS Neuron and accelerating deep learning models for GPU / CPU / Inferentia architectures.
- 2+ years of relevant work or research experience in system performance analysis, compiler optimizations (CUDA, OpenCL).
- Strong understanding with one or more of the following technologies : XLA, TVM, MLIR, LLVM, deep learning models and algorithms, and deep learning framework design.
- Knowledge of professional software engineering & best practices for full software development life cycle, including coding standards, software architectures, code reviews, source control management, continuous deployments, testing, and operational excellence