建模与仿真工程师
Modeling & Simulation Engineer
## **关于 Positron AI**
Positron AI 专注于开发定制的硬件系统以加速 AI 推理。这些推理系统在性能和效率方面相比传统的 GPU 系统有显著提升，在每美元性能和每瓦性能方面都有优势。Positron 的目标是打造世界上最好的 AI 推理系统。
**职位概述**
构建仿真和建模基础设施，帮助 Positron AI 在芯片可用之前开发推理系统。你将开发分析性能模型、功能和时序模拟器，以及支持架构探索和真实固件、运行时和主机软件工作负载的虚拟平台。该职位结合了计算机架构、软件工程和定量分析：选择合适的细节级别，验证模型行为，并使仿真成为日常工程中的可靠工具。
### **主要职责**
**仿真架构与建模**
- 开发推理工作负载的分析模型，评估延迟、吞吐量、利用率以及计算、内存和通信瓶颈。
- 构建并扩展加速器计算、内存、数据传输、互连和控制处理器的功能和时序模型，明确模型精度与仿真速度之间的权衡。
- 将加速器模型与虚拟平台和主机接口模型集成，以支持真实软件工作负载和连接的仿真环境。
- 通过可重复的实验比较当前和提议的架构，并说明每个结果背后的假设和限制。
**生产对齐与工程基础设施**
- 保持模拟器接口与生产固件、运行时和主机软件一致，以便工作负载、诊断和测试可以延续到硬件。
- 针对参考实现和可用硬件结果验证数值行为和时序假设；构建自动化回归覆盖，区分功能正确性与性能准确性。
- 通过配置工具、检查点、确定性执行、跟踪、可观测性和测试自动化来提高模拟器速度和易用性。
**跨职能协作**
- 与硬件和软件架构师紧密合作，讨论命令语义、调度假设和模型接口，确保模拟器行为与架构规范保持一致。
查看英文原文
## **About Positron AI**
Positron AI specializes in developing custom hardware systems to accelerate AI inference. These inference systems offer significant performance and efficiency gains over traditional GPU-based systems, delivering advantages in both performance per dollar and performance per watt. Positron exists to create the world's best AI inference systems.
**Role Overview**
Build the simulation and modeling infrastructure that helps Positron AI develop inference systems before silicon is available. You will develop analytical performance models, functional and timing simulators, and virtual platforms that support architecture exploration and realistic firmware, runtime, and host software workloads. The role combines computer architecture, software engineering, and quantitative analysis: choosing the right level of detail, validating model behavior, and making simulation a reliable part of everyday engineering.
### **Key Responsibilities**
**Simulation Architecture and Modeling**
- Develop analytical models of inference workloads to evaluate latency, throughput, utilization, and compute, memory, and communication bottlenecks.
- Build and extend functional and timing models of accelerator compute, memory, data movement, interconnects, and control processors, with explicit tradeoffs between model fidelity and simulation speed.
- Integrate accelerator models with virtual platforms and host interface models to support realistic software workloads and connected simulation environments.
- Compare current and proposed architectures through reproducible experiments, and communicate the assumptions and limits behind each result.
**Production Alignment and Engineering Infrastructure**
- Keep simulator interfaces aligned with production firmware, runtime, and host software so workloads, diagnostics, and tests can carry forward to hardware.
- Validate numerical behavior and timing assumptions against reference implementations and available hardware results; build automated regression coverage that distinguishes functional correctness from performance accuracy.
- Improve simulator speed and usability through configuration tools, checkpointing, deterministic execution, tracing, observability, and test automation.
**Cross-Functional Collaboration**
- Collaborate closely with hardware and software architects on command semantics, scheduling assumptions, and model interfaces, keeping simulator behavior consistent with architectural specifications and execution software.
- Partner with compiler, firmware, runtime, performance, and test engineers to turn models into practical development and qualification environments.
### **Required Qualifications**
- Strong software engineering skills in a systems language such as C++ or Rust, with experience building simulators, performance models, emulators, or comparable systems tools.
- Deep understanding of processors, memory hierarchies, interconnects, concurrency, and the effects of data movement and contention on performance.
- Experience designing and validating models with explicit goals for fidelity, determinism, and execution speed, and sound judgment about what to model or abstract.
- Strong applied mathematics and quantitative reasoning skills, with the ability to turn workload behavior into useful models and experiments.
- Ability to integrate complex software systems, debug interactions across components, and build maintainable tools that other engineers can use effectively.
- Clear technical writing and communication, including documenting assumptions, explaining discrepancies, and presenting architecture tradeoffs.
### **Preferred Qualifications**
- Experience with one or more of: SystemC, Arm Fast Models (LISA), QEMU, gem5, or custom event-driven or cycle-level simulation frameworks.
- Experience modeling AI inference workloads, including tensor operations, quantization, memory traffic, and token-generation performance.
- Familiarity with embedded Linux, firmware or runtime integration, device models, PCIe, or direct memory access.
- Experience with performance traces and visualization tools such as Perfetto, model calibration, distributed simulation, or test orchestration at scale.
- Experience with functional programming in languages such as Haskell or Lean, and the ability to work with domain-specific command languages used to model and control accelerator execution.
#### **Leveling & Scope**
While this role is currently posted at a specific level, we are a growth oriented organization and are open to hiring at a more senior level for the right candidate. Please note that this job description serves as a focused but generalized overview of the role; specific responsibilities and impact expectations will be tailored to the experience and seniority of the final hire.
### **What Success Looks Like**
In the first 12 months, you will help establish a representative end-to-end inference workload running through the simulation environment, with useful traces and documented fidelity and limitations. Software teams will use the models in regular development and qualification, and at least one architecture decision will be informed by a reproducible, validated simulation or modeling experiment.
### **Why Join Us?**
- Build foundational simulation infrastructure that shapes current and future generations of inference accelerators.
- Work at the intersection of architecture, firmware, and software on hard, high impact technical problems with direct influence on real hardware decisions.
### **Compensation & Benefits**
The base salary range for this role is $200,000 – $300,000.
Please note that the figures provided represent the base salary range only and do not include other elements of our total compensation package, equity, or comprehensive benefits.
At Positron AI, we value the unique expertise each candidate brings. While the range above reflects our typical expectation for the position, we reserve the flexibility to exceed this range for candidates whose specialized skills, significant experience, or unique qualifications fall outside the standard scope of the role. Final offers are determined based on a variety of factors, including internal equity, and individual impact.
### **Benefits & Perks**
We want you to do your best work and feel confident that you and your family are taken care of. That means comprehensive coverage, real time to rest, and support for your future.
**Health and wellness**
- Fully company paid medical, dental, and vision insurance for you and your dependents
- Company paid life and disability coverage, with voluntary options to add more
- Supplemental hospital, critical illness, and accident coverage available
**Time off and flexibility**
- Unlimited paid time off, we encourage everyone to truly unplug and recharge
- 13 paid company holidays
- Remote first culture with a company provided computer and home office setup
**Compensation and future**
- Competitive salary and equity
- 401(k) with company matching, eligible from day one
### **Visa Support**
This position is open to candidates currently authorized to work in the U.S. We cannot provide new visa sponsorship for this role but are open to facilitating H-1B visa transfers for eligible candidates.
**Equal Opportunity Employer.** If you're excited about the role but don't meet every bullet, we'd still love to hear from you.