Senior Engineer 2: Inference Optimizations

Join the Sora Team at OpenAIThe Sora team is at the forefront of developing multimodal capabilities within OpenAI’s foundational models. We are a dynamic blend of research and product development, committed to integrating sophisticated multimodal functionalities into our AI offerings. Our focus is on delivering solutions that are not only reliable and intuit…

Join fal as we revolutionize the generative-media infrastructure landscape. Our mission is to enhance model inference performance, enabling creative experiences on an unprecedented scale. We are seeking a Staff Technical Lead for Inference & ML Performance, an individual who possesses a unique blend of deep technical knowledge and strategic foresight. In this pivotal role, you will lead a talented team dedicated to building and optimizing cutting-edge inference systems. If you're ready to influence the future of inference performance in a fast-paced and rapidly growing environment, we want to hear from you.Why This Role MattersIn this role, you will play a crucial part in shaping the future of fal’s inference engine, ensuring that our generative models consistently deliver outstanding performance. Your contributions will directly affect our capacity to swiftly provide innovative creative solutions to a diverse clientele, from individual creators to global brands.Your ResponsibilitiesDefine and steer the technical direction, guiding your team across various domains including kernels, applied performance, ML compilers, and distributed inference to develop high-performance solutions.

About Our TeamJoin the Future of Computing Research team at OpenAI, an innovative applied research group within the Consumer Devices division. Our mission is to pioneer new methods and models that contribute to our overarching goal of developing Artificial General Intelligence (AGI) for the betterment of humanity.Role OverviewAs the Inference Technical Lead, you will collaborate with world-class machine learning researchers and top-notch design talents to push the boundaries of model capabilities. This position is stationed in San Francisco, CA, offering a hybrid work model that includes 4 days in the office, along with relocation assistance for new hires.Key ResponsibilitiesAssess and select silicon platforms, including GPUs, NPUs, and specialized accelerators, for the deployment of OpenAI models on-device and at the edge.Collaborate closely with research teams to co-design model architectures that satisfy real-world constraints such as latency, memory, power, and bandwidth.Conduct system performance analyses to identify trade-offs in model design, memory hierarchy, compute throughput, and hardware capabilities.Work hand-in-hand with hardware vendors and internal infrastructure teams to launch new accelerators, ensuring efficient execution of transformer workloads.Lead a team of engineers in implementing the low-level inference stack, encompassing kernel development and runtime systems.Navigate challenges to transform emerging research capabilities into scalable solutions.Ideal Candidate ProfileProven experience in evaluating or deploying workloads on GPUs, NPUs, or other specialized accelerators.Strong understanding of transformer model performance characteristics, including attention mechanisms, KV-cache behaviors, and memory bandwidth requirements.Experience designing or optimizing high-performance computing systems, such as inference engines, distributed runtimes, or hardware-aware ML pipelines.Background in building or leading teams focused on low-level performance-critical software, including CUDA kernels, compilers, or ML runtimes.Demonstrated ability to thrive in a fast-paced, innovative environment.

At Inferact, we are on a mission to establish vLLM as the premier AI inference engine, revolutionizing AI progress by making inference both more accessible and efficient. Our founding team consists of the original creators and key maintainers of vLLM, positioning us uniquely at the nexus of cutting-edge models and advanced hardware.Role OverviewWe are seeking a passionate inference runtime engineer eager to explore and expand the frontiers of LLM and diffusion model serving. As models evolve and grow in complexity with new architectures like mixture-of-experts and multimodal designs, the demand for innovative solutions in our inference engine intensifies. This role places you at the heart of vLLM, where you will enhance model execution across a variety of hardware platforms and architectures. Your contributions will have a direct influence on the future of AI inference.

Role overview Anthropic seeks a Technical Program Manager to support the Cloud Inference team. This position centers on steering technical projects that influence the development of cloud inference solutions. The role is located in either San Francisco, CA or New York City, NY. What you will do Oversee complex initiatives that move Anthropic’s cloud inference technologies forward Collaborate with engineers and partner teams to ensure delivery of dependable solutions Organize and synchronize work across different functions to achieve project objectives and deadlines

At Gimlet Labs, we are pioneering the development of the first heterogeneous neocloud designed specifically for AI workloads. As the demand for AI systems surges, traditional homogeneous infrastructures face critical limits in power, capacity, and cost. Our innovative platform effectively decouples AI workloads from their hardware foundations, intelligently partitioning tasks and orchestrating them to the most suitable hardware for optimal performance and efficiency. This strategy fosters heterogeneous systems that span multiple vendors and generations, including cutting-edge accelerators, enabling significant enhancements in performance and cost-effectiveness at scale.In addition to this foundational work, Gimlet is establishing a robust neocloud for agentic workloads. Our clients benefit from deploying and managing their workloads via stable, production-ready APIs, without the need to navigate hardware selection or performance optimization intricacies.We collaborate with foundation labs, hyperscalers, and AI-native companies to drive real production workloads capable of scaling to gigawatt-class AI datacenters.We are currently seeking a Member of Technical Staff specializing in ML systems and inference. In this pivotal role, you will be responsible for designing and constructing inference systems that facilitate the execution of complete models in real production environments. You will operate at the intersection of model architecture and system performance to ensure that inference processes are swift, predictable, and scalable.This position is perfect for engineers with a deep understanding of modern model execution and a passion for optimizing latency, throughput, and memory utilization across the entire inference lifecycle.

At Magic, we are driven by our mission to develop safe Artificial General Intelligence (AGI) that propels humanity forward in addressing the most critical challenges. We firmly believe that the future of safe AGI lies in automating research and code generation, allowing us to enhance models and tackle alignment issues more effectively than humans alone can manage. Our innovative approach combines cutting-edge pre-training, domain-specific reinforcement learning (RL), ultra-long context, and efficient inference-time computation to realize this vision.Position OverviewAs a Software Engineer within the Inference & RL Systems team, you will play a pivotal role in designing and managing the distributed systems that enable our models to function seamlessly in production, supporting extensive post-training workflows.This position operates at the intersection of model execution and distributed infrastructure, focusing on systems that influence inference latency, throughput, stability, and the reliability of RL and post-training training loops.Our long-context models impose significant execution demands, including KV-cache scaling, managing memory constraints for lengthy sequences, batching strategies, long-horizon trajectory rollouts, and ensuring consistent throughput under real-world workloads. You will be responsible for the infrastructure that ensures both production inference and large-scale RL iterations are efficient and dependable.Key ResponsibilitiesCraft and scale high-performance inference serving systems.Optimize KV-cache management, batching methods, and scheduling processes.Enhance throughput and latency for long-context tasks.Develop and sustain distributed RL and post-training infrastructure.Boost reliability across rollout, evaluation, and reward pipelines.Automate fault detection and recovery mechanisms for serving and RL systems.Analyze and eliminate performance bottlenecks across GPU, networking, and storage components.Collaborate with Kernel and Research teams to ensure alignment between execution systems and model architecture.QualificationsSolid foundation in software engineering and distributed systems.Proven experience in building or managing large-scale inference or training systems.In-depth understanding of GPU execution constraints and memory trade-offs.Experience troubleshooting performance issues in production machine learning systems.Capability to analyze system-level trade-offs between latency, throughput, and cost.

About Our TeamAt OpenAI, our Foundations team is dedicated to examining how model behavior evolves as we scale up models, data, and computing resources. We meticulously analyze the relationships between model architecture, optimization strategies, and training datasets to inform the design and training of next-generation models.About the PositionAs a Team Lead in Research Inference, you will be instrumental in constructing systems that empower advanced AI models to operate efficiently at scale. Your role lies at the crossroads of model research and systems engineering, where you will translate innovative architectural concepts into high-performance inference systems, clearly illustrating the trade-offs in performance, memory usage, and scalability.Your contributions will significantly shape model design, evaluation, and iteration processes across our research organization. By developing and refining high-performance inference infrastructures, you will provide researchers with the tools necessary to explore new ideas while understanding their computational and systems implications.This position does not involve serving products; instead, it supports research through a focus on performance, accuracy, and realism, ensuring that our AI research is firmly rooted in scalable solutions.ResponsibilitiesDesign and develop optimized inference runtimes for large-scale AI models, emphasizing efficiency, reliability, and scalability.Take ownership of optimizing core execution processes, including model execution, memory management, batching, and scheduling.Enhance and expand distributed inference across multiple GPUs, focusing on parallelism, communication patterns, and runtime coordination.Implement and refine critical inference operators and kernels based on real-world workloads.Collaborate closely with research teams to ensure accurate and efficient support for new model architectures within inference systems.Identify and resolve performance bottlenecks through comprehensive profiling, benchmarking, and low-level debugging.Contribute to the observability, correctness, and reliability of large-scale AI systems.Ideal Candidate ProfileExperience in developing production-level inference systems, beyond just training and executing models.Proficient in GPU-centric performance engineering, including managing memory behavior and understanding latency/throughput trade-offs.Strong analytical skills and familiarity with performance profiling tools.

OverviewAt Pulse, we are revolutionizing the way data infrastructure operates by addressing the critical challenge of accurately extracting structured information from intricate documents on a large scale. Our innovative document understanding technique merges intelligent schema mapping with advanced extraction models, outperforming traditional OCR and parsing methods.Located in the heart of San Francisco, we are a dynamic team of engineers dedicated to empowering Fortune 100 enterprises, YC startups, public investment firms, and growth-stage companies. Backed by top-tier investors, we are rapidly expanding our footprint in the industry.What sets our technology apart is our sophisticated multi-stage architecture, which includes:Specialized models for layout understanding and component detectionLow-latency OCR models designed for precise extractionAdvanced algorithms for reading-order in complex document structuresProprietary methods for table structure recognition and parsingFine-tuned vision-language models for interpreting charts, tables, and figuresIf you possess a strong passion for the convergence of computer vision, natural language processing, and data infrastructure, your contributions at Pulse will significantly impact our clients and help shape the future of document intelligence.

Join Cartesia as an Inference EngineerAt Cartesia, our vision is to create the next evolution of AI: an interactive, omnipresent intelligence that operates seamlessly across all environments. Currently, even the most advanced models struggle to continuously analyze a year's worth of audio, video, and text data—comprising 1 billion text tokens, 10 billion audio tokens, and 1 trillion video tokens—much less perform these tasks on-device.We are at the forefront of developing the model architectures that will make this a reality. Our founding team, who met as PhD candidates at the Stanford AI Lab, pioneered State Space Models (SSMs), a groundbreaking framework for training efficient, large-scale foundation models. Our talented team merges deep expertise in model innovation and systems engineering with a design-focused product engineering approach, enabling us to build and launch state-of-the-art models and user experiences.Supported by leading investors such as Index Ventures and Lightspeed Venture Partners, along with contributions from Factory, Conviction, A Star, General Catalyst, SV Angel, Databricks, and others, we are fortunate to be guided by numerous exceptional advisors and over 90 angel investors from diverse industries, including some of the world’s foremost experts in AI.About the RoleWe are actively seeking an Inference Engineer to propel our mission of creating real-time multimodal intelligence.Your ImpactDevelop and implement a low-latency, scalable, and dependable model inference and serving stack for our innovative foundation models utilizing Transformers, SSMs, and hybrid models.Collaborate closely with our research team and product engineers to efficiently deliver our product suite in a fast, cost-effective, and reliable manner.Construct robust inference infrastructure and monitoring systems for our product offerings.Enjoy substantial autonomy in shaping our products and directly influencing how cutting-edge AI is integrated across diverse devices and applications.What You BringAt Cartesia, we prioritize strong engineering skills due to the complexity and scale of the challenges we tackle.Proficient engineering skills with a comfort level in navigating intricate codebases, and a commitment to producing clean, maintainable code.Experience in developing large-scale distributed systems with strict performance, reliability, and observability requirements.Proven technical leadership, capable of executing and delivering results from zero to one amidst uncertainty.A background in or experience with inference pipelines, machine learning, and generative models.

P-1285 About This Role Join Databricks as a Staff Software Engineer specializing in GenAI inference, where you will spearhead the architecture, development, and optimization of the inference engine that powers the Databricks Foundation Model API. Your role will be crucial in bridging cutting-edge research with real-world production requirements, ensuring exceptional throughput, minimal latency, and scalable solutions. You will work across the entire GenAI inference stack, including kernels, runtimes, orchestration, memory management, and integration with various frameworks and orchestration systems. What You Will Do Take full ownership of the architecture, design, and implementation of the inference engine, collaborating on a model-serving stack optimized for large-scale LLM inference. Work closely with researchers to integrate new model architectures or features, such as sparsity, activation compression, and mixture-of-experts into the engine. Lead comprehensive optimization efforts focused on latency, throughput, memory efficiency, and hardware utilization across GPUs and other accelerators. Establish and uphold standards for building and maintaining instrumentation, profiling, and tracing tools to identify performance bottlenecks and drive optimizations. Design scalable solutions for routing, batching, scheduling, memory management, and dynamic loading tailored to inference workloads. Guarantee reliability, reproducibility, and fault tolerance in inference pipelines, including capabilities for A/B testing, rollbacks, and model versioning. Collaborate cross-functionally to integrate with federated and distributed inference infrastructure, ensuring effective orchestration across nodes, load balancing, and minimizing communication overhead. Foster collaboration with cross-functional teams, including platform engineers, cloud infrastructure, and security/compliance professionals. Represent the team externally through benchmarks, whitepapers, and contributions to open-source projects. What We Look For A BS/MS/PhD in Computer Science or a related discipline. A solid software engineering background with 6+ years of experience in performance-critical systems. A proven ability to own complex system components and influence architectural decisions from conception to execution. A deep understanding of ML inference internals, including attention mechanisms, MLPs, recurrent modules, quantization, and sparse operations. Hands-on experience with CUDA, GPU programming, and essential libraries (cuBLAS, cuDNN, NCCL, etc.). A strong foundation in distributed systems design, including RPC frameworks, queuing, RPC batching, sharding, and memory partitioning. Demonstrated proficiency in diagnosing and resolving performance bottlenecks across multiple layers (kernel, memory, networking, scheduler).

About Liquid AIBorn from the innovation of MIT CSAIL, Liquid AI is at the forefront of developing general-purpose AI systems that operate seamlessly across various deployment platforms, including data center accelerators and on-device hardware. Our solutions prioritize low latency, minimal memory consumption, privacy, and reliability. We collaborate with leading enterprises in sectors such as consumer electronics, automotive, life sciences, and financial services. As we experience rapid growth, we seek extraordinary talent to join our mission.The OpportunityJoin our Edge Inference team, where we transform Liquid Foundation Models into highly optimized machine code for resource-limited devices such as smartphones, laptops, Raspberry Pis, and smartwatches. As key contributors to llama.cpp, we establish the infrastructure necessary for efficient on-device AI. You will collaborate closely with our technical lead to tackle complex challenges that demand a profound understanding of machine learning architectures and hardware constraints. This role offers high ownership, allowing your code to be deployed in production environments and directly influence model performance on real devices.While San Francisco and Boston are preferred, we welcome applicants from other locations.

Cohere builds and deploys advanced AI models used by developers and enterprises. These models support applications like content generation, semantic search, retrieval-augmented generation (RAG), and intelligent agents. The team’s work aims to make AI more accessible and practical for real-world use. Each person at Cohere plays a direct role in strengthening the models and increasing their value for clients. The company values practical outcomes and continuous improvement, focusing on delivering reliable technology to users. The team includes researchers, engineers, designers, and professionals from a wide range of backgrounds. Cohere believes that diverse perspectives help create better products. The company welcomes those interested in shaping the future of AI to join its mission.

ABOUT BASETENAt Baseten, we empower the leading AI companies of today, including Cursor, Notion, OpenEvidence, Abridge, Clay, Gamma, and Writer, by providing essential inference capabilities. Our unique blend of applied AI research, adaptable infrastructure, and intuitive developer tools enables innovators at the cutting edge of AI to seamlessly transition advanced models into production. With our recent success in securing a $300M Series E funding round, backed by notable investors such as BOND, IVP, Spark Capital, Greylock, and Conviction, we're on an exciting growth trajectory. Join our team and contribute to the platform that engineers rely on to launch AI-driven products.THE ROLEAs an Applied AI Inference Engineer at Baseten, you'll collaborate closely with clients to design, develop, and implement high-performance AI applications using our platform. You will guide customers through the entire process, from initial concept to deployment, transforming vague business objectives into dependable, observable solutions that meet defined quality, latency, and cost metrics.This position is ideal for innovative engineers eager to gain insight into how modern organizations scale AI adoption. You will thrive if you enjoy a multifaceted role that intersects product development, software engineering, performance optimization, and direct customer engagement.It’s essential to note that this position requires hands-on coding and software development, while also encompassing elements of product management, technical customer success, and pre-sales engineering.EXAMPLE INITIATIVESExplore insights from our Forward Deployed Engineering team through these blog posts: Forward Deployed Engineering on the frontier of AIThe fastest, most accurate Whisper transcriptionDeploy production-ready model servers from Docker imagesDeploy custom ComfyUI workflows as APIs...

About Our TeamJoin the Inference team at OpenAI, where we leverage cutting-edge research and technology to deliver exceptional AI products to consumers, enterprises, and developers. Our mission is to empower users to harness the full potential of our advanced AI models, enabling unprecedented capabilities. We prioritize efficient and high-performance model inference while accelerating research advancements.About the RoleWe are seeking a passionate Software Engineer to optimize some of the world's largest and most sophisticated AI models for deployment in high-volume, low-latency, and highly available production and research environments.Key ResponsibilitiesCollaborate with machine learning researchers, engineers, and product managers to transition our latest technologies into production.Work closely with researchers to enable advanced research initiatives through innovative engineering solutions.Implement new techniques, tools, and architectures that enhance the performance, latency, throughput, and effectiveness of our model inference stack.Develop tools to identify bottlenecks and instability sources, designing and implementing solutions for priority issues.Optimize our code and Azure VM fleet to maximize every FLOP and GB of GPU RAM available.You Will Excel in This Role If You:Possess a solid understanding of modern machine learning architectures and an intuitive grasp of performance optimization strategies, especially for inference.Take ownership of problems end-to-end, demonstrating a willingness to acquire any necessary knowledge to achieve results.Bring at least 5 years of professional software engineering experience.Have or can quickly develop expertise in PyTorch, NVidia GPUs, and relevant optimization software stacks (such as NCCL, CUDA), along with HPC technologies like InfiniBand, MPI, and NVLink.Have experience in architecting, building, monitoring, and debugging production distributed systems, with bonus points for working on performance-critical systems.Have successfully rebuilt or significantly refactored production systems multiple times to accommodate rapid scaling.Are self-driven, enjoying the challenge of identifying and addressing the most critical problems.

About Our TeamJoin OpenAI’s dynamic Inference team, where we empower the deployment of cutting-edge AI models, including our renowned GPT models, advanced Image Generation capabilities, and Whisper, across diverse platforms. Our mission is to ensure these models are not only high-performing and scalable but also available for real-world applications. Collaborating closely with our Research team, we’re committed to bringing the next generation of AI innovations to fruition. As a compact, agile team, we prioritize delivering an exceptional developer experience while continuously pushing the frontiers of artificial intelligence.As we expand our focus into multimodal inference, we are building the necessary infrastructure to support models that process images, audio, and other non-text modalities. This work involves tackling diverse model sizes and interactions, managing complex input/output formats, and ensuring seamless collaboration between product and research teams.About The RoleWe are seeking a passionate Software Engineer to aid in the large-scale deployment of OpenAI’s multimodal models. You will join a small yet impactful team dedicated to creating robust, high-performance infrastructure for real-time audio, image, and various multimodal workloads in production environments.This position is inherently collaborative; you will work directly with researchers who develop these models and with product teams to define novel interaction modalities. Your contributions will enable users to generate speech, interpret images, and engage with models in innovative ways that extend beyond traditional text-based interactions.Key Responsibilities:Design and implement advanced inference infrastructure for large-scale multimodal models.Optimize systems for high-throughput and low-latency processing of image and audio inputs and outputs.Facilitate the transition of experimental research workflows into dependable production services.Engage closely with researchers, infrastructure teams, and product engineers to deploy state-of-the-art capabilities.Contribute to systemic enhancements, including GPU utilization, tensor parallelism, and hardware abstraction layers.You May Excel In This Role If You:Have a proven track record of building and scaling inference systems for large language models or multimodal architectures.Possess experience with GPU-based machine learning workloads and a solid understanding of the performance dynamics associated with large models, particularly with intricate data types like images or audio.Thrive in a fast-paced, experimental environment and enjoy collaborating with cross-functional teams to drive impactful results.

Join Anthropic as an Engineering Manager to lead our innovative Cloud Inference team utilizing AWS technologies. In this pivotal role, you will drive efforts to enhance the efficiency and scalability of our cloud systems while ensuring robust performance and reliability. Your leadership will inspire a talented team of engineers to solve complex challenges, implement best practices, and foster a culture of continuous improvement.

About Our TeamThe Inference team at OpenAI is dedicated to translating our cutting-edge research into accessible, transformative technology for consumers, enterprises, and developers. By leveraging our advanced AI models, we enable users to achieve unprecedented levels of innovation and productivity. Our primary focus lies in enhancing model inference efficiency and accelerating progress in research through optimized inference capabilities.About the RoleWe are seeking talented engineers to expand and optimize OpenAI's inference infrastructure, specifically targeting emerging GPU platforms. This role encompasses a wide range of responsibilities from low-level kernel optimization to high-level distributed execution. You will collaborate closely with our research, infrastructure, and performance teams to ensure seamless operation of our largest models on cutting-edge hardware.This position offers a unique opportunity to influence and advance OpenAI’s multi-platform inference capabilities, with a strong emphasis on optimizing performance for AMD accelerators.Your Responsibilities Include:Overseeing the deployment, accuracy, and performance of the OpenAI inference stack on AMD hardware.Integrating our internal model-serving infrastructure (e.g., vLLM, Triton) into diverse GPU-backed systems.Debugging and optimizing distributed inference workloads across memory, network, and compute layers.Validating the correctness, performance, and scalability of model execution on extensive GPU clusters.Collaborating with partner teams to design and optimize high-performance GPU kernels for accelerators utilizing HIP, Triton, or other performance-centric frameworks.Working with partner teams to develop, integrate, and fine-tune collective communication libraries (e.g., RCCL) to parallelize model execution across multiple GPUs.Ideal Candidates Will:Possess experience in writing or porting GPU kernels using HIP, CUDA, or Triton, with a strong focus on low-level performance.Be familiar with communication libraries like NCCL/RCCL, understanding their importance in high-throughput model serving.Have experience with distributed inference systems and be adept at scaling models across multiple accelerators.Enjoy tackling end-to-end performance challenges across hardware, system libraries, and orchestration layers.Be eager to join a dynamic, agile team focused on building innovative infrastructure from the ground up.

Join our innovative team at Anthropic as a Software Engineer specializing in Cloud Inference Safeguards. In this role, you will play a crucial part in developing and enhancing the systems that ensure the robustness and security of our cloud-based inference services. You will collaborate with cross-functional teams to design, implement, and maintain scalable solutions that meet our high standards for reliability and performance.