Togetherai

<h3><strong>About The Role </strong></h3>
<p>We're looking for a software engineer, reliability-minded but still a builder, to own what happens in production to the Kubernetes-native control plane that provisions and runs our GPU inference fleet. The platform team builds the manifest-driven API and controllers that let the inference team declare what they need, a cluster, a model deployment, a capacity change, and get it reconciled underneath. Your job is to keep that promise true in the real world: triaging InfiniBand issues when prefill and decode nodes stall exchanging KV cache over RDMA, diagnosing hero-model and LoRA-serving faults that don't show up in generic health checks, and extending the automated self-healing system so each new failure mode becomes a permanent fix instead of a recurring page.</p>
<p>The core value we're after is keeping the decoupling real even when things break, and keeping the fleet efficient without that efficiency becoming a new source of instability. You'll own SLOs and error budgets for inference serving, lead incident response, and turn what postmortems teach you into actual changes in the controllers and automation, not just updated runbooks. You'll also work the reliability side of capacity efficiency, since a mistuned defragmentation or bin-packing pass is itself a production risk. Strong candidates have hands-on production-reliability experience on large-scale distributed or GPU-serving systems, are comfortable troubleshooting InfiniBand/RDMA and Kubernetes controller/CRD-based platforms, and think about GPU infrastructure, and the tools that keep it reliable, as software to be engineered.</p>
<p>A product mindset applies: the on-call experience, the clarity of an alert, and the judgment built into an automated remediation are things you ship and iterate on.</p>
<h3><strong>Responsibilities</strong></h3>
<ul>
<li>Operate the provisioning state machine in production: own the health of the host lifecycle pipeline, discovery, inference bring-up, GPU driver/CUDA validation, health checks, decommission/RMA, triaging stuck or failed state transitions at fleet scale and feeding real failure patterns back into how the states and transitions are designed.</li>
<li>Triage and extend automated self-healing: when the node-scanning and remediation platform misses something, a new failure mode, an InfiniBand issue in disaggregated prefill/decode serving, a GPU Xid variant, a hero-model or LoRA-specific fault, diagnose it live, drive the fix to production, and turn it into a new detection/remediation plugin so it's automatic the next time.</li>
<li>Own end-to-end reliability of the pipeline: hold the SLOs and error budgets for provisioning and fleet health, lead incident response when idempotency, retry, rollback, or drift assumptions break down in practice, and drive the systemic fix, not just the hotfix.</li>
<li>Partner with the inference/ML platform team from the production side: be the one who knows how real cluster topologies, interconnects, and scheduling constraints actually fail in the field, and feed those failure modes back as first-class requirements into the platform's abstractions.</li>
<li>Work alongside machine learning researchers, engineers, and product managers to bring our latest technologies into production, enabling advanced research through awesome engineering.</li>
<li>Introduce new techniques, tools, and architecture that improve the performance, latency, throughput, and efficiency of our model inference stack.</li>
<li>Build tools to give us visibility into our bottlenecks and sources of instability, then design and implement solutions to address the highest-priority issues.</li>
<li>Lead reliability cross-functionally: build coalition across infra, ML platform, and product teams around shared reliability priorities, manage stakeholders through incidents and roadmap trade-offs, and represent the reliability point of view in planning so it shapes decisions before they ship, not after an outage.</li>
</ul>
<h3><strong>Requirements</strong></h3>
<ul>
<li>Experience operating a provisioning or host-lifecycle pipeline in production, comfortable troubleshooting stuck or failed state transitions (discovery, bring-up, GPU driver/CUDA validation, health checks, decommission/RMA) at fleet scale, and translating real failure patterns back into improvements to the state machine’s design.</li>
<li>Deep experience owning production reliability for large-scale distributed systems; GPU/ML-serving infrastructure experience is a strong plus.</li>
<li>Hands-on experience troubleshooting InfiniBand/RDMA networking issues, ideally in a multi-node training or disaggregated-inference context — comfortable reading IB port counters and subnet manager logs, not just application metrics.</li>
<li>Familiarity with the operational realities of multi-tenant model serving (LoRA/adapter-based serving) and with running differentiated SLOs across model pools of different priority.</li>
<li>A demonstrated track record of eliminating operational toil through automation and tooling you built yourself, not just process or documentation.</li>
<li>Hands-on experience with Kubernetes controller/CRD patterns, reconciliation loops, and node-level health/remediation systems.</li>
<li>Comfort with event-driven architectures, backpressure/circuit-breaking design, and treating observability (metrics, logs, traces) as something built into a system, not added after an incident.</li>
<li>A software-engineering mindset first: you write and ship production code, and you think about GPU infrastructure as software to be engineered, not a fleet to be manually operated.</li>
</ul>
<h3><strong>About Together AI</strong></h3>
<p>Together AI, the AI Native Cloud, is purpose-built for AI engineers. AI application developers get high-performance inference that scales reliably, fine-tuning and reinforcement learning for creating frontier-level specialized models, and pre-training at massive scale for fully custom intelligence, all around a marketplace of leading open models that teams can run, adapt, and own. Trusted by Cursor, Decagon, ElevenLabs, Salesforce, and Zoom, Together serves 400+ trillion tokens a month.</p>
<h3><strong>Compensation</strong></h3>
<p>We offer competitive compensation, startup equity, health insurance and other competitive benefits. The US base salary range for this full-time position is: $250,000 - $320,000 + equity + benefits. Our salary ranges are determined by location, level and role. Individual compensation will be determined by experience, skills, and job-related knowledge.</p>
<h3><strong>Equal Opportunity</strong></h3>
<p>Together AI is an Equal Opportunity Employer and is proud to offer equal employment opportunity to everyone regardless of race, color, ancestry, religion, sex, national origin, sexual orientation, age, citizenship, marital status, disability, gender identity, veteran status, and more.</p>
<p>Please see our privacy policy at <a href="https://www.together.ai/privacy" rel="noopener noreferrer ugc" target="_blank">https://www.together.ai/privacy</a>. </p>

(Staff / Senior Staff) AI Infra Engineer, Inference Infrastructure

<h3><span><strong>About the Role</strong></span></h3>
<p><span>The Speculative Decoding team at Together AI builds the draft models and algorithms that let large language models generate faster without changing their outputs. Our speculators run in production on frontier open models such as DeepSeek, Kimi and GLM. Our work sits at the intersection of model training, inference algorithms and serving systems, and we co-design all three to lower the cost and latency of generation.</span></p>
<p><span>As a research intern, you will dive into open questions in speculative decoding: new drafter architectures and verification schemes, knowledge distillation and training recipes that align draft and target models, and drafters that adapt as traffic shifts. You will train and fine-tune draft models at scale on dedicated GPU clusters, and test your ideas against realistic serving workloads, including long-context and agentic traffic.</span></p>
<p><span>Projects aim to answer important research questions in speculative decoding, enabling faster generation on frontier models, drafters that keep pace with changing workloads, and robust, reproducible evaluation under real traffic.</span></p>
<h3><span>Past Research</span></h3>
<p><span>Speculative decoding research by Together researchers, interns and academic collaborators includes:</span></p>
<ul>
<li><span><a href="https://openreview.net/forum?id=aL1Wnml9Ef" target="_blank" rel="noopener noreferrer ugc">Speculative Speculative Decoding</a> (ICLR 2026)</span></li>
<li><span><a href="https://arxiv.org/abs/2602.06932" target="_blank" rel="noopener noreferrer ugc">When RL Meets Adaptive Speculative Training: A Unified Training-Serving System</a> (ICML 2026)</span></li>
<li><span><a href="https://arxiv.org/abs/2511.13841" target="_blank" rel="noopener noreferrer ugc">Beat the Long Tail: Distribution-Aware Speculative Decoding for RL Training</a> (MLSys 2026)</span></li>
</ul>
<h3><span><strong>Responsibilities</strong></span></h3>
<ul>
<li><span>Research and prototype new efficient speculative decoding algorithms and draft model architectures</span></li>
<li><span>Train, fine-tune and distill draft models, and design rigorous experiments to validate hypotheses.</span></li>
<li><span>Implement optimized version for serving engines to show efficacy in real world workloads</span></li>
<li><span>Communicate the plans, progress, and results of projects to the broader team</span></li>
<li><span>Share findings through technical write-ups and blog posts, and help bring successful ideas into production</span></li>
</ul>
<h3><span><strong>Requirements</strong></span></h3>
<ul>
<li><span>Currently pursuing a Master's or Ph.D. degree in Computer Science, Electrical Engineering, or a related field</span></li>
<li><span>Strong knowledge of Machine Learning and Deep Learning fundamentals</span></li>
<li><span>Hands-on experience training or fine-tuning large models with distributed training frameworks (e.g., Megatron-LM, FSDP)</span></li>
<li><span>Familiarity with modern model architectures, such as Mixture-of-Experts and hybrid attention</span></li>
</ul>
<h3><span>Preferred Qualifications</span></h3>
<ul>
<li><span>Prior research experience in speculative decoding, efficient inference, or efficient machine learning</span></li>
<li><span>Publications at leading conferences in machine learning or systems (e.g., NeurIPS, ICML, ICLR, MLSys)</span></li>
<li><span>Experience with knowledge distillation, post-training, or reinforcement learning</span></li>
<li><span>Familiarity with LLM serving engines (e.g., vLLM, SGLang, TensorRT-LLM)</span></li>
<li><span>Contributions to open-source machine learning projects</span></li>
</ul>
<h3><span><strong>About Together AI</strong></span></h3>
<p><span>Together AI, the AI Native Cloud, is purpose-built for AI engineers. AI application developers get high-performance inference that scales reliably, fine-tuning and reinforcement learning for creating frontier-level specialized models, and pre-training at massive scale for fully custom intelligence, all around a marketplace of leading open models that teams can run, adapt, and own. Trusted by Cursor, Decagon, ElevenLabs, Salesforce, and Zoom, Together serves 400+ trillion tokens a month.</span></p>
<h3><span><strong>Internship Program Details</strong></span></h3>
<p><span>Our internship program runs 12 to 14 weeks, giving you the opportunity to work alongside industry-leading engineers and researchers across multiple teams. This cohort's internship dates span May 17th to August 6th or June 14th to September 3rd.</span></p>
<h3><span><strong>Compensation</strong></span></h3>
<p><span>We offer competitive compensation, housing stipends, and other competitive benefits. The estimated US hourly rate for this role is $58 - $70 an hour. Our hourly rates are determined by location, level and role. Individual compensation will be determined by experience, skills, and job-related knowledge.</span></p>
<h3><span><strong>Equal Opportunity</strong></span></h3>
<p><span>Together AI is an Equal Opportunity Employer and is proud to offer equal employment opportunity to everyone regardless of race, color, ancestry, religion, sex, national origin, sexual orientation, age, citizenship, marital status, disability, gender identity, veteran status, and more.</span></p>
<p><span>For more information about our privacy policy, please visit:<a href="https://www.together.ai/privacy" rel="noopener noreferrer ugc" target="_blank"> https://www.together.ai/privacy</a></span></p>
<h3><br /><br /><br /></h3>

Research Intern, Frontier Speculative Decoding (Summer 2027)

<h3><span><strong>About the Role</strong></span></h3>
<p><span>The Speculative Decoding team at Together AI builds the draft models and algorithms that let large language models generate faster without changing their outputs. Our speculators run in production on frontier open models such as DeepSeek, Kimi and GLM. Our work sits at the intersection of model training, inference algorithms and serving systems, and we co-design all three to lower the cost and latency of generation.</span></p>
<p><span>As a research intern, you will dive into open questions in speculative decoding: new drafter architectures and verification schemes, knowledge distillation and training recipes that align draft and target models, and drafters that adapt as traffic shifts. You will train and fine-tune draft models at scale on dedicated GPU clusters, and test your ideas against realistic serving workloads, including long-context and agentic traffic.</span></p>
<p><span>Projects aim to answer important research questions in speculative decoding, enabling faster generation on frontier models, drafters that keep pace with changing workloads, and robust, reproducible evaluation under real traffic.</span></p>
<h3><span>Past Research</span></h3>
<p><span>Speculative decoding research by Together researchers, interns and academic collaborators includes:</span></p>
<ul>
<li><span><a href="https://openreview.net/forum?id=aL1Wnml9Ef" target="_blank" rel="noopener noreferrer ugc">Speculative Speculative Decoding</a> (ICLR 2026)</span></li>
<li><span><a href="https://arxiv.org/abs/2602.06932" target="_blank" rel="noopener noreferrer ugc">When RL Meets Adaptive Speculative Training: A Unified Training-Serving System</a> (ICML 2026)</span></li>
<li><span><a href="https://arxiv.org/abs/2511.13841" target="_blank" rel="noopener noreferrer ugc">Beat the Long Tail: Distribution-Aware Speculative Decoding for RL Training</a> (MLSys 2026)</span></li>
</ul>
<h3><span><strong>Responsibilities</strong></span></h3>
<ul>
<li><span>Research and prototype new efficient speculative decoding algorithms and draft model architectures</span></li>
<li><span>Train, fine-tune and distill draft models, and design rigorous experiments to validate hypotheses.</span></li>
<li><span>Implement optimized version for serving engines to show efficacy in real world workloads</span></li>
<li><span>Communicate the plans, progress, and results of projects to the broader team</span></li>
<li><span>Share findings through technical write-ups and blog posts, and help bring successful ideas into production</span></li>
</ul>
<h3><span><strong>Requirements</strong></span></h3>
<ul>
<li><span>Currently pursuing a Master's or Ph.D. degree in Computer Science, Electrical Engineering, or a related field</span></li>
<li><span>Strong knowledge of Machine Learning and Deep Learning fundamentals</span></li>
<li><span>Hands-on experience training or fine-tuning large models with distributed training frameworks (e.g., Megatron-LM, FSDP)</span></li>
<li><span>Familiarity with modern model architectures, such as Mixture-of-Experts and hybrid attention</span></li>
</ul>
<h3><span>Preferred Qualifications</span></h3>
<ul>
<li><span>Prior research experience in speculative decoding, efficient inference, or efficient machine learning</span></li>
<li><span>Publications at leading conferences in machine learning or systems (e.g., NeurIPS, ICML, ICLR, MLSys)</span></li>
<li><span>Experience with knowledge distillation, post-training, or reinforcement learning</span></li>
<li><span>Familiarity with LLM serving engines (e.g., vLLM, SGLang, TensorRT-LLM)</span></li>
<li><span>Contributions to open-source machine learning projects</span></li>
</ul>
<h3><span><strong>About Together AI</strong></span></h3>
<p><span>Together AI, the AI Native Cloud, is purpose-built for AI engineers. AI application developers get high-performance inference that scales reliably, fine-tuning and reinforcement learning for creating frontier-level specialized models, and pre-training at massive scale for fully custom intelligence, all around a marketplace of leading open models that teams can run, adapt, and own. Trusted by Cursor, Decagon, ElevenLabs, Salesforce, and Zoom, Together serves 400+ trillion tokens a month.</span></p>
<h3><span><strong>Internship Program Details</strong></span></h3>
<p><span>Our internship program runs 12 to 14 weeks, giving you the opportunity to work alongside industry-leading engineers and researchers across multiple teams. This cohort's internship dates span January 4th to April 9th.</span></p>
<h3><span><strong>Compensation</strong></span></h3>
<p><span>We offer competitive compensation, housing stipends, and other competitive benefits. The estimated US hourly rate for this role is $58 - $70 an hour. Our hourly rates are determined by location, level and role. Individual compensation will be determined by experience, skills, and job-related knowledge.</span></p>
<h3><span><strong>Equal Opportunity</strong></span></h3>
<p><span>Together AI is an Equal Opportunity Employer and is proud to offer equal employment opportunity to everyone regardless of race, color, ancestry, religion, sex, national origin, sexual orientation, age, citizenship, marital status, disability, gender identity, veteran status, and more.</span></p>
<p><span>For more information about our privacy policy, please visit:<a href="https://www.together.ai/privacy" rel="noopener noreferrer ugc" target="_blank"> https://www.together.ai/privacy</a></span></p>
<h3><br /><br /><br /></h3>

Research Intern, Frontier Speculative Decoding (Winter 2027)

<h3><strong>About the Role</strong></h3>
<p>Together AI is building the AI Native Cloud, an end-to-end platform for the full generative AI lifecycle, combining the fastest LLM inference engine with state-of-the-art GPU cloud infrastructure. The Together Cloud team builds the [Together GPU Clusters](https://www.together.ai/gpu-clusters) flagship IaaS product that provides high-performance, AI-ready GPU clusters through a self-serve cloud console, along with the virtualized infrastructure layer powering Together's inference, RL, and fine-tuning products.</p>
<p>As a Staff Software Engineer focusing on AI Compute in the Together Cloud org, you will set technical direction for and build major components of the next generation AI cloud platform – a highly available, global cloud infrastructure with cutting-edge virtualization of the latest ML hardware: GB300s/VRs, BlueField DPUs, InfiniBand and dual/quad-plane RoCEv2 fabrics. That virtualized computing platform powers our own SaaS products – inference, RL, and fine-tuning – and serves external cloud customers through self-serve offerings such as on-demand/reserved Kubernetes/Slurm clusters, across dozens of data centers and hundreds of thousands of GPUs.</p>
<p>This is an architect-and-build role. Fully automated bootstrapping of GPU data centers, high-performance virtualization of GPU compute and DC networking without compromising isolation or portability, and fault-tolerant decentralized control planes — you'll set the architecture for these across our global and in-DC services, and be a key owner of the hardest parts, in the code as well as the design. Your designs will span the IaaS layer of a greenfield Vera Rubin data center up to the global management plane that schedules capacity across all of them. At this level the job is as much leverage as code: the standards you set and the engineers you grow decide how fast the rest of Together Cloud ships.</p>
<h3><strong>Responsibilities</strong></h3>
<ul>
<li><strong>Own the GPU and network virtualization stack: </strong>the hypervisor, kernel, and SDN work that keeps GPU compute and DC networking high-performance, portable, and strongly isolated across heterogeneous hardware.</li>
<li><strong>Own the in-DC IaaS layer:</strong> architect and roadmap the services, Kubernetes operators, and libraries that provision and manage compute, storage, and networks in our data centers — VMs, parallel filesystems, VPCs, and InfiniBand partitions; lead its build-out for a new Vera Rubin data center with thousands of GPUs, from hardware bring-up to customer-facing API.</li>
<li><strong>Design the GPU scheduling and global management plane:</strong> the distributed control plane behind on-demand and reserved clusters across dozens of data centers, including the systems that scale per-cluster limits and automate the onboarding of new capacity.</li>
<li><strong>Architect monitoring and automated remediation for fault tolerance: </strong>the strategy for automated detection, isolation, and recovery of failed nodes that keeps distributed pretraining and large-scale inference running through hardware failures.</li>
<li><strong>Set technical direction across teams:</strong> lead design reviews, resolve cross-cutting architectural disagreements, unblock cross-team dependencies and integration risks, and define the standards other engineers build against — measured in cluster reliability, time-to-first-GPU on new capacity, and quality at scale.</li>
<li><strong>Grow the team:</strong> mentor senior and junior engineers, deepen the team's expertise in virtualization, DC networking, and GPU infrastructure, and help raise the hiring bar for Together Cloud.</li>
<li><strong>Set the engineering bar:</strong> create the testing frameworks, tools, and developer documentation that make our systems robust and usable by other teams, and shape the core, open-source Together AI platform.</li>
</ul>
<p>To be successful you'll need to be deeply technical and an excellent communicator — expert software development fundamentals, deep systems knowledge and troubleshooting instincts, and the leadership and diplomacy skills to align teams that don't report to you. Much of this work starts ambiguous, and we expect you to define the scope yourself and drive it to production.</p>
<h3><strong>Requirements</strong></h3>
<ul>
<li>7+ years of professional software development experience, with expert-level proficiency in at least one backend language (Golang desired), writing high-performance, well-tested, production-quality code.</li>
<li>Track record of owning the architecture of large distributed systems from blank page to production at scale, including the judgment calls that could not be reversed cheaply.</li>
<li>Deep experience building and operating globally distributed, high-performance microservice architectures across one or more cloud providers (AWS, Azure, GCP).</li>
<li>Expert systems knowledge across compute, networking, and storage — including concurrency, memory management, performant I/O, and scale at a global level.</li>
<li>Demonstrated technical leadership beyond your own commits: mentoring senior engineers, leading design reviews, and driving alignment across teams that do not report to you.</li>
<li>Excellent communication and diplomacy skills — able to write design docs that settle arguments, and to work effectively with technical and non-technical stakeholders.</li>
<li>Experience building and operating reliable, customer-facing production systems at scale, and owning the infrastructure automation (Terraform, Ansible), observability (Prometheus, Grafana), and CI/CD (GitHub Actions, ArgoCD) that keep them healthy.</li>
</ul>
<h3><strong>Preferred Qualifications (not must haves)</strong></h3>
<ul>
<li>Deep Kubernetes internals experience, such as implementing non-trivial Kubernetes operators, device/storage/network plugins, custom schedulers, or patches to Kubernetes itself</li>
<li>Deep experience with VMs/hypervisors, such as QEMU/KVM, cloud-hypervisor, VFIO, virtio, PCIE passthrough, Kubevirt, SR-IOV</li>
<li>Deep experience with DC networking tech + solutions, such as VLAN, VXLAN, VPN, VPC, OVS/OVN</li>
<li>Experience with Cluster API or similar</li>
<li>Experience working on high-performance compute, networking, and/or storage</li>
<li>Experience virtualizing GPUs and/or InfiniBand</li>
<li>Experience building IaaS or PaaS systems at scale</li>
<li>Experience with DPUs/SmartNICs</li>
<li>GPU programming, NCCL, CUDA knowledge</li>
</ul>
<h3><strong>About Together AI</strong></h3>
<p>Together AI, the AI Native Cloud, is purpose-built for AI engineers. AI application developers get high-performance inference that scales reliably, fine-tuning and reinforcement learning for creating frontier-level specialized models, and pre-training at massive scale for fully custom intelligence, all around a marketplace of leading open models that teams can run, adapt, and own. Trusted by Cursor, Decagon, ElevenLabs, Salesforce, and Zoom, Together serves 400+ trillion tokens a month.</p>
<h3><strong>Equal Opportunity</strong></h3>
<p>Together AI is an Equal Opportunity Employer and is proud to offer equal employment opportunity to everyone regardless of race, color, ancestry, religion, sex, national origin, sexual orientation, age, citizenship, marital status, disability, gender identity, veteran status, and more.</p>
<div>
<div>
<div>
<p>Please see our privacy policy at <a href="https://www.together.ai/privacy" rel="noopener noreferrer ugc" target="_blank">https://www.together.ai/privacy</a>  </p>
</div>
</div>
</div>

Staff Software Engineer - AI Compute, Together Cloud

<h3><strong>About the Role</strong></h3>
<p>Together AI is building the AI Native Cloud, an end-to-end platform for the full generative AI lifecycle, combining the fastest LLM inference engine with state-of-the-art GPU cloud infrastructure. The Together Cloud team builds the [Together GPU Clusters](https://www.together.ai/gpu-clusters) flagship IaaS product that provides high-performance, AI-ready GPU clusters through a self-serve cloud console, along with the virtualized infrastructure layer powering Together's inference, RL, and fine-tuning products.</p>
<p>As a Staff Software Engineer focusing on AI Compute in the Together Cloud org, you will set technical direction for and build major components of the next generation AI cloud platform – a highly available, global cloud infrastructure with cutting-edge virtualization of the latest ML hardware: GB300s/VRs, BlueField DPUs, InfiniBand and dual/quad-plane RoCEv2 fabrics. That virtualized computing platform powers our own SaaS products – inference, RL, and fine-tuning – and serves external cloud customers through self-serve offerings such as on-demand/reserved Kubernetes/Slurm clusters, across dozens of data centers and hundreds of thousands of GPUs.</p>
<p>This is an architect-and-build role. Fully automated bootstrapping of GPU data centers, high-performance virtualization of GPU compute and DC networking without compromising isolation or portability, and fault-tolerant decentralized control planes — you'll set the architecture for these across our global and in-DC services, and be a key owner of the hardest parts, in the code as well as the design. Your designs will span the IaaS layer of a greenfield Vera Rubin data center up to the global management plane that schedules capacity across all of them. At this level the job is as much leverage as code: the standards you set and the engineers you grow decide how fast the rest of Together Cloud ships.</p>
<h3><strong>Responsibilities</strong></h3>
<ul>
<li><strong>Own the GPU and network virtualization stack: </strong>the hypervisor, kernel, and SDN work that keeps GPU compute and DC networking high-performance, portable, and strongly isolated across heterogeneous hardware.</li>
<li><strong>Own the in-DC IaaS layer:</strong> architect and roadmap the services, Kubernetes operators, and libraries that provision and manage compute, storage, and networks in our data centers — VMs, parallel filesystems, VPCs, and InfiniBand partitions; lead its build-out for a new Vera Rubin data center with thousands of GPUs, from hardware bring-up to customer-facing API.</li>
<li><strong>Design the GPU scheduling and global management plane:</strong> the distributed control plane behind on-demand and reserved clusters across dozens of data centers, including the systems that scale per-cluster limits and automate the onboarding of new capacity.</li>
<li><strong>Architect monitoring and automated remediation for fault tolerance: </strong>the strategy for automated detection, isolation, and recovery of failed nodes that keeps distributed pretraining and large-scale inference running through hardware failures.</li>
<li><strong>Set technical direction across teams:</strong> lead design reviews, resolve cross-cutting architectural disagreements, unblock cross-team dependencies and integration risks, and define the standards other engineers build against — measured in cluster reliability, time-to-first-GPU on new capacity, and quality at scale.</li>
<li><strong>Grow the team:</strong> mentor senior and junior engineers, deepen the team's expertise in virtualization, DC networking, and GPU infrastructure, and help raise the hiring bar for Together Cloud.</li>
<li><strong>Set the engineering bar:</strong> create the testing frameworks, tools, and developer documentation that make our systems robust and usable by other teams, and shape the core, open-source Together AI platform.</li>
</ul>
<p>To be successful you'll need to be deeply technical and an excellent communicator — expert software development fundamentals, deep systems knowledge and troubleshooting instincts, and the leadership and diplomacy skills to align teams that don't report to you. Much of this work starts ambiguous, and we expect you to define the scope yourself and drive it to production.</p>
<h3><strong>Requirements</strong></h3>
<ul>
<li>7+ years of professional software development experience, with expert-level proficiency in at least one backend language (Golang desired), writing high-performance, well-tested, production-quality code.</li>
<li>Track record of owning the architecture of large distributed systems from blank page to production at scale, including the judgment calls that could not be reversed cheaply.</li>
<li>Deep experience building and operating globally distributed, high-performance microservice architectures across one or more cloud providers (AWS, Azure, GCP).</li>
<li>Expert systems knowledge across compute, networking, and storage — including concurrency, memory management, performant I/O, and scale at a global level.</li>
<li>Demonstrated technical leadership beyond your own commits: mentoring senior engineers, leading design reviews, and driving alignment across teams that do not report to you.</li>
<li>Excellent communication and diplomacy skills — able to write design docs that settle arguments, and to work effectively with technical and non-technical stakeholders.</li>
<li>Experience building and operating reliable, customer-facing production systems at scale, and owning the infrastructure automation (Terraform, Ansible), observability (Prometheus, Grafana), and CI/CD (GitHub Actions, ArgoCD) that keep them healthy.</li>
</ul>
<h3><strong>Preferred Qualifications (not must haves)</strong></h3>
<ul>
<li>Deep Kubernetes internals experience, such as implementing non-trivial Kubernetes operators, device/storage/network plugins, custom schedulers, or patches to Kubernetes itself</li>
<li>Deep experience with VMs/hypervisors, such as QEMU/KVM, cloud-hypervisor, VFIO, virtio, PCIE passthrough, Kubevirt, SR-IOV</li>
<li>Deep experience with DC networking tech + solutions, such as VLAN, VXLAN, VPN, VPC, OVS/OVN</li>
<li>Experience with Cluster API or similar</li>
<li>Experience working on high-performance compute, networking, and/or storage</li>
<li>Experience virtualizing GPUs and/or InfiniBand</li>
<li>Experience building IaaS or PaaS systems at scale</li>
<li>Experience with DPUs/SmartNICs</li>
<li>GPU programming, NCCL, CUDA knowledge</li>
</ul>
<h3><strong>About Together AI</strong></h3>
<p>Together AI is a research-driven artificial intelligence company. We believe open and transparent AI systems will drive innovation and create the best outcomes for society, and together we are on a mission to significantly lower the cost of modern AI systems by co-designing software, hardware, algorithms, and models. We have contributed to leading open-source research, models, and datasets to advance the frontier of AI, and our team has been behind technological advancement such as FlashAttention, Hyena, FlexGen, and RedPajama. We invite you to join a passionate group of researchers in our journey in building the next generation AI infrastructure.</p>
<h3><strong>Compensation</strong></h3>
<p>We offer competitive compensation, startup equity, health insurance, and other benefits, as well as flexibility in terms of remote work. The US base salary range for this full-time position is: $260,000 - $300,000 + equity + benefits. Our salary ranges are determined by location, level and role. Individual compensation will be determined by experience, skills, and job-related knowledge.</p>
<h3><strong>Equal Opportunity</strong></h3>
<p>Together AI is an Equal Opportunity Employer and is proud to offer equal employment opportunity to everyone regardless of race, color, ancestry, religion, sex, national origin, sexual orientation, age, citizenship, marital status, disability, gender identity, veteran status, and more.</p>
<div>
<div>
<div>
<p>Please see our privacy policy at <a href="https://www.together.ai/privacy" rel="noopener noreferrer ugc" target="_blank">https://www.together.ai/privacy</a>  </p>
</div>
</div>
</div>

Togetherai

Open jobs at Togetherai

(Staff / Senior Staff) AI Infra Engineer, Inference Infrastructure

Research Intern, Frontier Speculative Decoding (Summer 2027)

Research Intern, Frontier Speculative Decoding (Winter 2027)

Staff Software Engineer - AI Compute, Together Cloud

Staff Software Engineer - AI Compute, Together Cloud

Working for Togetherai?

Featured companies