דרושים » הנדסה » Software Engineer - Computation Kernels and Performance

As a Drivers & Virtualization Engineer, you will own the host-facing software boundary. Your mission is to ensure our hardware is accessible, secure, and performant across diverse environments-ranging from bare-metal deployments to complex virtualized and multi-device cloud infrastructures. You will focus on building robust kernel-level interfaces where correctness, isolation, and long-term operational safety are paramount.

Responsibilities
Develop and maintain high-performance Linux kernel drivers for proprietary accelerator.

Implement communication protocols and virtual device interfaces to bridge hardware and software.

Manage low-level hardware interactions including PCIe, DMA, MMIO, and interrupt handling.

Enable seamless virtualized access to accelerators using QEMU and KVM.

Design and implement robust device sharing, passthrough, and isolation mechanisms.

Ensure the security of accelerator operations through strict memory protection and resource isolation.

Debug complex failures that span the entire stack: from hardware and kernel to hypervisor and guest OS.

Support the scaling of our architecture from single-device setups to large-scale multi-device deployments.

We are looking for a Software Development Engineer to own the design and implementation of our inference data plane. We build the software that makes large models run efficiently on custom hardware - spanning model execution, memory management, data movement, and serving integration.
Our work covers the full inference path: integrating serving engines with custom hardware, developing high-performance compute kernels, enabling efficient data movement, and driving models from early validation through production. We operate at frontier scale with large distributed models.
This is a ground-up effort with rapidly evolving hardware and software. We need an individual contributor who can write and optimize low-level code for custom hardware, validate model architectures end-to-end, build test and profiling infrastructure, and drive performance across the stack.

Key job responsibilities
- Develop and optimize compute kernels for a custom ML accelerator architecture, targeting production-level performance for large language model inference.
- Implement and validate LLM architectures end-to-end - from PyTorch model definition through distributed execution on custom hardware.
- Integrate custom accelerator backends into open-source ML serving frameworks (vLLM, PyTorch), including scheduler extensions, memory management, and model parallelism.
- Build and maintain test infrastructure for model correctness validation across CPU, GPU, simulator, and hardware targets.
- Profile and optimize inference workloads - identify bottlenecks, instrument critical paths, and drive latency and throughput improvements from simulation through hardware bringup.
- Own features end-to-end: from design through implementation, testing, and integration into the broader software stack.
- Contribute to CI/CD pipelines that gate model and kernel changes on correctness and performance regressions.

The MLIL DataPlane team is looking for a Senior Software Development Engineer to own the design and implementation of our inference data plane. We build the software that makes large models run efficiently on custom hardware - spanning model execution, memory management, data movement, and serving integration.
Our work covers the full inference path: integrating serving engines with custom hardware, developing high-performance compute kernels, enabling efficient data movement, and driving models from early validation through production. We operate at frontier scale with large distributed models.
This is a ground-up effort with rapidly evolving hardware and software. We need a senior IC who can write and optimize low-level code for custom hardware, validate model architectures end-to-end, build test and profiling infrastructure, and drive performance across the stack.

Key job responsibilities
- Develop and optimize compute kernels for a custom ML accelerator architecture, targeting production-level performance for large language model inference.
- Implement and validate LLM architectures (decoder-only, mixture-of-experts) end-to-end - from PyTorch model definition through distributed execution on custom hardware.
- Integrate custom accelerator backends into open-source ML serving frameworks (vLLM, PyTorch), including scheduler extensions, memory management, and model parallelism.
- Build and maintain test infrastructure for model correctness validation across CPU, GPU, simulator, and hardware targets.
- Profile and optimize inference workloads - identify bottlenecks, instrument critical paths, and drive latency and throughput improvements from simulation through hardware bringup.
- Own features end-to-end: from design through implementation, testing, and integration into the broader software stack.
- Contribute to CI/CD pipelines that gate model and kernel changes on correctness and performance regressions.
- Mentor engineers, drive design reviews, and raise the engineering bar across the team.

We are seeking a talented and experienced Software Engineer to design, build, and optimize high-performance distributed systems, core data engines, and backend infrastructure. This role requires deep system-level architecture understanding, the ability to handle large-scale clusters processing petabytes of data, and mastery of modern C/C++ and Linux environment internals.
Key Responsibilities
Design & Develop Core Components: Build, maintain, and optimize highly scalable, resilient distributed services, storage engines, or data-processing pipelines written in C/C++.
System Architecture & Resilience: Drive architectural discussions and implementations around high availability, data consistency, replication mechanisms, fault tolerance, and multi-node concurrency.
Performance Optimization: Optimize hot execution paths, low-level data structures, memory management, and I/O subsystems to guarantee high throughput and minimal latency.
Complex Debugging & Troubleshooting: Investigate and resolve intricate production issues spanning the application layer, distributed networking protocols, file systems, and operating system kernels.
End-to-End Ownership: Take full technical ownership of critical features-from ambiguous requirements and system design through implementation, rollout, and observability in production environments.
Teams
Storage Platform: Focuses on building a next-generation distributed storage platform handling petabytes of data across large clusters specifically designed to power AI, enterprise, and analytics workloads. Handling everything that touches the hardware and operating system aspects in a software defined storage system
Data Path: Focuses on engineering a highly distributed, latency-critical Hot I/O Data Path and Element Store engine. This role is responsible for the ingestion, state-of-the-art compression, encoding, and retrieval of multi-protocol data (files and objects) under massive concurrency and ultra-low latency requirements.
Database: Focuses deeply on core relational database internals, specifically designing low-level storage engines, B-Tree/LSM-Tree data structures, MVCC concurrency control, and query execution planners.
Kernel: Focuses on the lowest software layers, emphasizing Linux Kernel development and block-level storage/file system engineering.
Protocols: Focuses strictly on engineering high-concurrency data/metadata paths that replicate external AWS S3 object-storage behavior and correctness under heavy retry and failover pressure.
Cloud: Focuses on cloud-native storage deployment, adapting and scaling complex high-availability storage infrastructure across major hyper-scaler cloud environments (AWS, Azure, GCP).
Compute Kafka: Focuses on distributed event-streaming and messaging platforms, specifically building a high-scale, exactly-once broker compatible with the Apache Kafka wire protocol.

We are looking for talented Software Development Engineers to join our teams in Tel Aviv or Haifa and work on low-level software development. You will be building the foundational software layers that sit closest to the hardware - enabling high performance, reliability, and efficiency for AWS services at an unprecedented scale.

This is an opportunity to work in a deeply technical environment where your code runs on custom-designed silicon, directly impacting the performance and cost of cloud infrastructure used by millions of customers worldwide.

Design, develop, and optimize low-level software including device drivers, firmware, and hardware abstraction layers.

Collaborate closely with hardware design teams to define hardware-software interfaces and bring-up new silicon.

Write performance-critical systems software in C/C++ that operates under strict latency and resource constraints.

Debug complex hardware-software interactions.

Contribute to system architecture decisions and define software design patterns for next-gen chip features.

Develop and maintain automated test frameworks to validate functionality and performance at scale.

Participate in code reviews and drive engineering best practices within the team.

The MLIL DataPlane team is looking for a Software Development Engineer to own the design and implementation of our inference data plane. We build the software that makes large models run efficiently on custom hardware - spanning model execution, memory management, data movement, and serving integration.
Our work covers the full inference path: integrating serving engines with custom hardware, developing high-performance compute kernels, enabling efficient data movement, and driving models from early validation through production. We operate at frontier scale with large distributed models.
This is a ground-up effort with rapidly evolving hardware and software. We need an individual contributor who can write and optimize low-level code for custom hardware, validate model architectures end-to-end, build test and profiling infrastructure, and drive performance across the stack.

Key job responsibilities
- Develop and optimize compute kernels for a custom ML accelerator architecture, targeting production-level performance for large language model inference.
- Implement and validate LLM architectures end-to-end - from PyTorch model definition through distributed execution on custom hardware.
- Integrate custom accelerator backends into open-source ML serving frameworks (vLLM, PyTorch), including scheduler extensions, memory management, and model parallelism.
- Build and maintain test infrastructure for model correctness validation across CPU, GPU, simulator, and hardware targets.
- Profile and optimize inference workloads - identify bottlenecks, instrument critical paths, and drive latency and throughput improvements from simulation through hardware bringup.
- Own features end-to-end: from design through implementation, testing, and integration into the broader software stack.
- Contribute to CI/CD pipelines that gate model and kernel changes on correctness and performance regressions.

At our company, we build AI-powered vision systems that enhance safety and decision-making for some of the worlds largest vessels.
Our platform processes live video streams from multiple onboard cameras to provide real-time situational awareness, detecting and tracking marine objects, even in low visibility and highly congested environments. These systems directly support navigational decisions and help prevent collisions, reduce human error, and improve operational efficiency.
Our systems are already deployed across thousands of vessels and have processed hundreds of millions of nautical miles of real-world data, operating in unpredictable and safety-critical conditions.
This role sits at the intersection of AI and high-performance systems engineering, focused on solving real-world problems under strict constraints. You will work on systems where performance and reliability are critical and where improvements have a direct, measurable impact on real-world safety.
This is a senior, systems-focused role with end-to-end ownership over performance and reliability of production computer vision pipelines. You will define optimization strategies, identify bottlenecks across the system, and drive improvements under real-world constraints.
What youll do
Build and optimize real-time computer vision pipelines running on edge systems processing live maritime video streams (e.g, NVIDIA Jetson, Triton Inference Server)
Take models from research and turn them into production-ready, reliable components deployed on vessels
Profile and improve end-to-end system performance across: multi-camera video ingestion; preprocessing; inference; postprocessing
Identify and resolve bottlenecks across CPU, GPU, memory, and pipeline coordination
Make and justify tradeoffs between latency, accuracy, stability, and resource utilization
Design and implement robust data and inference pipelines (video -> model -> actionable output for crew)
Develop benchmarking and evaluation workflows to measure performance end-to-end and support release gating
Build and improve observability tools, including logging, monitoring, and debugging workflows for production systems
Define and maintain clear interfaces between research code and production systems
Work closely with research and backend teams to integrate new models into production systems
Continuously improve system efficiency and reliability under hardware and runtime constraints.

Required Software Engineer III, Cloud Networking
About the job
Our software engineers develop the next-generation technologies that change how billions of users connect, explore, and interact with information and one another. Our products need to handle information at massive scale, and extend well beyond web search. We're looking for engineers who bring fresh ideas from all areas, including information retrieval, distributed computing, large-scale system design, networking and data storage, security, artificial intelligence, natural language processing, UI design and mobile; the list goes on and is growing every day. As a software engineer, you will work on a specific project critical to our needs with opportunities to switch teams and projects as you and our fast-paced business grow and evolve. We need our engineers to be versatile, display leadership qualities and be enthusiastic to take on new problems across the full-stack as we continue to push technology forward.
This Team owns the core data plane infrastructure for all NAT-based products within the Andromeda network stack (e.g., private service connect (PSC) and Cloud network address translation (NAT). These products serve as critical entry and exit gateways that securely connect Cloud customers to their services and networks.
Responsibilities
Design, develop, test, and maintain high-performance software for our Cloud's core networking and secure connectivity platforms.
Manage complex scalability, resource efficiency, and performance optimization challenges to evolve our cloud networking datapath.
Collaborate with cross-functional engineering partners to define and build new network architectures and routing methods.
Manage individual project priorities, deadlines, and deliverables, ensuring high engineering velocity and robust code quality.
Participate in operational rotations, monitor production health, and troubleshoot packet latency or connectivity issues to keep our global systems healthy.

Help build an Always-On, low-overhead GPU profiling service that runs in production, scales across cluster environments, and delivers actionable insights for ML workloads. You will be hands-on delivering our profiling solutions across system software, drivers, and CUDA to make profiling continuously available and reliable.

What youll be doing:

Develop low-overhead, high-reliability implementations in C/C++, with bounded CPU/memory budgets.

Lead end-to-end feature delivery spanning user-mode components, driver/platform layers, and performance counter/trace providers.

Establish profiling models that integrate with existing ML/AI workflows (e.g., PyTorch/XLA) to turn low-level signals into actionable insights.

Required Staff Design Engineer, Networking, Cloud
About the job
Be part of a team that pushes boundaries, developing custom silicon solutions that power the future of our direct-to-consumer products. You'll contribute to the innovation behind products loved by millions worldwide. Your expertise will shape the next generation of hardware experiences, delivering unparalleled performance, efficiency, and integration.
In this role, you will be part of a team developing application-specific integrated circuits (ASICs) used to accelerate networking in data centers. You will have multiple responsibilities in areas such as project definition, design, and implementation. You will participate in the design, architecture, documentation, and implementation of the next generation of data center accelerators.You will also be responsible for performance analysis for a networking stack using your knowledge.
We prioritize security, efficiency, and reliability across everything we do - from developing our latest TPUs to running a global network, while driving towards shaping the future of hyperscale computing. Our global impact spans software and hardware, including Clouds Vertex AI, the leading AI platform for bringing Gemini models to enterprise customers.
Responsibilities
Lead an ASIC subsystem.
Understand how it interacts with software and other ASIC subsystems to implement data center networks.
Define hardware/software interfaces. Write micro architecture and design specifications.
Define efficient micro-architecture and block partitioning/interfaces and flows.
Collaborate closely with software, verification, and physical design stakeholders to ensure the designs are complete, correct, and performant.