系统架构师 AI/ML 基础设施
Systems Architect AI/ML Infrastructure
## **公司简介**
Deepgram 是支撑新兴万亿美元语音 AI 经济的领先平台，提供实时 API 用于语音转文本（STT）、文本转语音（TTS），并可大规模构建生产级语音代理。超过 20 万名开发者和 1,300 多家机构正在构建“由 Deepgram 提供支持”的语音产品，包括 Twilio、Cloudflare、Sierra、Decagon、Vapi、Daily、Cresta、Granola 和 Jack in the Box。Deepgram 的语音原生基础模型通过云 API 或自托管及本地软件进行访问，具有无与伦比的准确性、低延迟和成本效率。由领先的全球投资者和战略合作伙伴牵头的 C 轮融资支持，Deepgram 已处理超过 5 万年的音频，并转录了超过 1 万亿个单词。世界上没有哪家公司比 Deepgram 更了解语音。
## **公司运营节奏**
在 Deepgram，我们期望具备 AI 优先的思维模式——AI 的使用和熟悉度不是可选的，而是我们运作、创新和衡量绩效的核心。
在 Deepgram 工作的每位成员都应积极使用和尝试先进的 AI 工具，甚至将它们融入日常工作中。我们衡量 AI 在交付结果中的应用效果，持续且富有创意地利用最新 AI 能力是这里成功的关键。候选人应能够快速采用新模型和新模式，将 AI 整合到工作流程中，并不断推动这些技术的边界。
此外，我们以 AI 的速度前进。变化迅速，你可以期待你的日常工作也以同样快的速度演变。如果你对实验、适应、临场思考和不断学习不感兴趣，或者你寻求的是高度规范化的传统朝九晚五工作，那么这可能不是适合你的职位。
### **机会介绍**
Deepgram 的基础设施涵盖裸金属 GPU 集群、多云部署和全球边缘节点——所有这些都在大规模实时语音 AI 服务的同时，支持大规模模型训练。作为系统架构师，你将负责实现这一切的端到端基础设施架构。你将设计用于生产推理和研究训练工作负载的计算、存储和网络系统，构建平衡性能与成本的多云策略，并创建可扩展的突发性基础设施，以适应 Deepgram 快速增长的规模。
查看英文原文
## **Company Overview**
Deepgram is the leading platform underpinning the emerging trillion-dollar Voice AI economy, providing real-time APIs for speech-to-text (STT), text-to-speech (TTS), and building production-grade voice agents at scale. More than 200,000 developers and 1,300+ organizations build voice offerings that are ‘Powered by Deepgram’, including Twilio, Cloudflare, Sierra, Decagon, Vapi, Daily, Cresta, Granola, and Jack in the Box. Deepgram’s voice-native foundation models are accessed through cloud APIs or as self-hosted and on-premises software, with unmatched accuracy, low latency, and cost efficiency. Backed by a recent Series C led by leading global investors and strategic partners, Deepgram has processed over 50,000 years of audio and transcribed more than 1 trillion words. There is no organization in the world that understands voice better than Deepgram.
## **Company Operating Rhythm**
At Deepgram, we expect an AI-first mindset—AI use and comfort aren’t optional, they’re core to how we operate, innovate, and measure performance.
Every team member who works at Deepgram is expected to actively use and experiment with advanced AI tools, and even build your own into your everyday work. We measure how effectively AI is applied to deliver results, and consistent, creative use of the latest AI capabilities is key to success here. Candidates should be comfortable adopting new models and modes quickly, integrating AI into their workflows, and continuously pushing the boundaries of what these technologies can do.
Additionally, we move at the pace of AI. Change is rapid, and you can expect your day-to-day work to evolve just as quickly. This may not be the right role if you’re not excited to experiment, adapt, think on your feet, and learn constantly, or if you’re seeking something highly prescriptive with a traditional 9-to-5.
### **The Opportunity**
Deepgram's infrastructure spans bare metal GPU clusters, multi-cloud deployments, and global edge presence -- all serving real-time voice AI at massive scale while simultaneously powering large-scale model training. As a Systems Architect, you will own the end-to-end infrastructure architecture that makes this possible. You will design the compute, storage, and networking systems that serve both production inference and research training workloads, build multi-cloud strategies that balance performance with cost, and create burstable infrastructure that scales with Deepgram's rapidly growing demands. This is a senior technical leadership role where your architectural decisions shape the foundation everything at Deepgram runs on.
### **What You'll Do**
- Define and drive the end-to-end infrastructure architecture for Deepgram's AI/ML workloads across production inference and research training
- Design multi-cloud and hybrid infrastructure strategies that balance performance, reliability, cost, and vendor flexibility
- Architect compute orchestration systems that efficiently schedule and manage GPU and CPU workloads across heterogeneous infrastructure
- Design storage architectures that handle the massive datasets required for speech and audio ML -- from high-throughput training data pipelines to low-latency model serving
- Lead capacity planning across all infrastructure dimensions, modeling growth and ensuring Deepgram can scale ahead of demand
- Drive cost optimization and FinOps practices, identifying opportunities to reduce infrastructure spend without compromising performance or reliability
- Design burstable, elastic training infrastructure that can scale up for large training runs and scale down to minimize idle cost
- Architect research compute infrastructure that gives ML teams the resources they need while maintaining operational efficiency
- Establish architectural standards, design review processes, and technical documentation practices for infrastructure decisions
- Collaborate with engineering leadership to align infrastructure strategy with product roadmap and business objectives
- Evaluate emerging hardware, cloud services, and infrastructure technologies for potential adoption
### **You'll Love This Role If You**
- Think in systems -- you naturally see the connections between compute, storage, network, and how they interact under load
- Are motivated by designing infrastructure that operates at the intersection of real-time production systems and large-scale ML training
- Enjoy making architectural trade-offs where cost, performance, reliability, and velocity are all in tension
- Want to work across the full infrastructure stack -- from bare metal and GPUs to cloud services and container orchestration
- Are excited about building cost-effective, burstable infrastructure that enables world-class AI research
- Like operating at a strategic level while staying technically deep enough to validate designs and debug complex issues
### **It's Important To Us That You Have**
- 7+ years of experience in infrastructure engineering, systems architecture, or a senior technical role focused on large-scale infrastructure
- Proven experience designing multi-cloud architectures spanning AWS and at least one other major cloud provider or on-premises environment
- Deep expertise in storage system design -- block, object, and file storage, including performance tuning for large-scale data workloads
- Strong experience with compute orchestration using Kubernetes, and an understanding of how to schedule diverse workloads efficiently
- Hands-on experience with GPU infrastructure -- procurement considerations, cluster design, driver and runtime management
- Track record of capacity planning and infrastructure scaling for high-growth environments
- Ability to communicate complex architectural decisions clearly to both technical and non-technical stakeholders
- Strong understanding of networking fundamentals as they relate to infrastructure architecture (see our Network Engineer role for the deep specialist)
### **It Would Be Great If You Had**
- Direct experience architecting infrastructure for ML training workloads -- distributed training, large dataset management, experiment infrastructure
- Background in cost optimization and FinOps practices for large-scale cloud and bare metal infrastructure
- Experience operating and managing bare metal infrastructure in colocation facilities
- Expertise in network architecture design, including high-bandwidth GPU interconnects and global traffic routing
- Experience with infrastructure modeling and simulation for capacity planning
- Familiarity with Slurm, Ray, or other HPC/ML job scheduling systems
- Understanding of power, cooling, and physical infrastructure considerations for GPU-dense deployments
_**Notice**: We're aware of individuals impersonating Deepgram recruiters. All legitimate Deepgram recruiting communication comes from an @_ [_deepgram.com_](http://deepgram.com) _email address. If you've received a message claiming to be Deepgram, please forward it to careers@deepgram.com._