{"id":72425,"title":"Working Student - LLM Inference Performance","company":"Tensordyne","city":"München","region":"Bayern","location":"Munich, Germany","work_model":"unknown","employment_type":"werkstudent","employment":["werkstudent"],"roles":["data_science"],"visa":"not_mentioned","relocation":"not_mentioned","posted_at":"2026-11-01T19:31:17+00:00","tools":["Python","C","C++","Rust","Embedded Systems","LLM"],"seniority":"junior","years_experience_min":null,"german_requirement":"not_mentioned","german_level":"unspecified","english_requirement":"required","english_level":"unspecified","team_language":"unknown","confidence":0.4,"method":"rules","language_badge":"Not stated","language_evidence":null,"requirements_snippet":"Currently enrolled in a Master's degree program in Computer Science, Electrical Engineering, Mathematics, or a related technical field.. Strong programming skills in Python and at least one of C, C++, or Rust.. Understanding of …","url":"https://job-boards.greenhouse.io/tensordyne/jobs/8926410002","source":"greenhouse","agency":null,"description_text":"About Tensordyne\n\nAI is transforming our world. It can perform cognitive functions that previously only humans could do, such as perceiving interactions across different modalities and environments - with the ability to quickly learn and then solve complex problems. Tensordyne is an AI system solution company that builds very high-performance, low-power generative AI inference systems. Our mission, through the creation of custom silicon, hardware and software, is to enable multimodal Generative AI inference acceleration at scale, with safe, sustainable, high-performance systems for our hyperscaler and neocloud data center customers. We are at the leading edge of advancing the latest research and product improvements for generative Al inference solutions that will make Al even more advantageous for compelling new generative AI applications.Tensordyne is a well funded, fast-paced startup company with headquarters in both Sunnyvale, CA, and Munich, Germany. We also have many talented team members working remotely across North America and Europe. We take care of our people and their families with comprehensive benefits, competitive compensation, flexible spending options, and recognition programs, because building category-defining technology starts with a healthy, supported team. Come join us as we shape the future of multimodal generative artificial intelligence!\n\nAbout the Role\n\nHow fast should an LLM building block run on our hardware? How fast does it actually run, and what would make it faster?\n\nAs a Working Student, LLM Inference Performance, you'll help answer these questions by working directly with our hardware and software engineering teams.\n\nYou'll write and optimize low-level kernels, benchmark performance on real silicon, and develop profiling tools that help engineers understand and improve LLM inference performance.\n\nYou'll gain hands-on experience with our Rust programming stack and Python SDK, use AI coding agents as part of your development workflow, and collaborate closely with engineers working on compilers, runtimes, and kernel optimization.\n\nThis is an opportunity to apply your academic knowledge to real-world AI hardware challenges while taking ownership of meaningful technical projects.\n\nWhat You'll Do\n\n•\nUnderstand our hardware: Learn how Tensordyne silicon processes data, manages memory, and synchronizes operations, and how key LLM components such as attention, MLP, and Mixture-of-Experts (MoE) routing map onto the architecture..\n\n•\nDevelop and optimize kernels: Implement, test, and tune low-level kernels to improve the performance of LLM inference workloads..\n\n•\nAnalyze performance: Build analytical and roofline performance models, compare predictions against simulator and real-silicon measurements, and investigate performance bottlenecks..\n\n•\nBuild benchmarking and profiling tools: Develop benchmark harnesses, profiling scripts, and trace visualizations that make performance easier to measure, understand, and improve..\n\n•\nWork with AI coding agents: Explore workflows where AI agents assist with kernel development, optimization, and benchmarking, and identify the tools and context they need to work effectively..\n\n•\nCollaborate across engineering teams: Work closely with compiler, runtime, and kernel engineers to translate performance insights into practical improvements..\n\nExample project: Develop a benchmarking and profiling harness for an LLM building block, such as decode attention, at two abstraction levels. Compare measured performance against theoretical estimates and make the tooling accessible to AI coding agents.\n\nWhat We're Looking For\n\n•\nCurrently enrolled in a Master's degree program in Computer Science, Electrical Engineering, Mathematics, or a related technical field..\n\n•\nStrong programming skills in Python and at least one of C, C++, or Rust..\n\n•\nUnderstanding of computer architecture, memory systems, bandwidth, and parallel computing through academic coursework or personal projects..\n\n•\nInterest in low-level programming, hardware performance, and optimization..\n\n•\nPractical experience using AI coding agents in development workflows, with the ability to evaluate and improve their output..\n\n•\nAn experimental mindset, with the ability to form hypotheses, analyze measurements, and communicate technical findings clearly..\n\n•\nCuriosity, initiative, and enthusiasm for learning new technologies..\n\nNice to Have\n\nExperience in one or more of the following areas would be a plus:\n\n•\nGPU or AI accelerator programming..\n\n•\nEmbedded systems or bare-metal programming..\n\n•\nPerformance profiling, benchmarking, or optimization..\n\n•\nLLM inference fundamentals and architectures..\n\nWhy Join Tensordyne?\n\n•\nWork on cutting-edge AI technology: Contribute to the development of next-generation AI inference hardware and software..\n\n•\nGain hands-on experience: Apply your university knowledge to real silicon and challenging engineering problems..\n\n•\nOwn meaningful projects: Build tools and solutions that directly support our engineering teams..\n\n•\nLearn from experienced engineers: Work alongside specialists in hardware architecture, compilers, runtimes, and performance optimization..\n\n•\nExplore AI-assisted development: Gain practical experience building with AI coding agents and shaping how they interact with advanced hardware systems..\n\n•\nMake a lasting impact: Help develop performance tools and workflows that will support both current and future generations of Tensordyne technology..\n\nTensordyne's culture was built on the following values\n• Put people first. We only succeed when our people succeed.\n• Ethics and integrity always. Be open, honest, and respectful of everyone.\n• Think Big. Be ambitious and have audacious goals.\n• Aim for excellence. Quality and excellence count in everything we do.\n• Own it and get it done. Results matter!\n• Make each person better together than they would be as an individual.\n• Embrace each other’s differences, and embrace that there will be differences.\n\nTensordyne is an equal opportunity employer. We believe diverse teams are better equipped to solve complex problems and build exceptional technology. All qualified applicants will receive consideration for employment without regard to age, color, gender identity or expression, marital status, national origin, disability, protected veteran status, race, religion, pregnancy, sexual orientation, or any other characteristic protected by applicable laws, regulations, and ordinances.\n\nA note to recruitment agencies: Please do not contact Tensordyne employees or leaders regarding this role. We do not accept unsolicited agency resumes and are not responsible for fees associated with unsolicited submissions.","responsibilities_text":"How fast should an LLM building block run on our hardware? How fast does it actually run, and what would make it faster?\nAs a Working Student, LLM Inference Performance, you'll help answer these questions by working directly with our hardware and software engineering teams.\nYou'll write and optimize low-level kernels, benchmark performance on real silicon, and develop profiling tools that help engineers understand and improve LLM inference performance.\nYou'll gain hands-on experience with our Rust programming stack and Python SDK, use AI coding agents as part of your development workflow, and collaborate closely with engineers working on compilers, runtimes, and kernel optimization.\nThis is an opportunity to apply your academic knowledge to real-world AI hardware challenges while taking ownership of meaningful technical projects.\n•\nUnderstand our hardware: Learn how Tensordyne silicon processes data, manages memory, and synchronizes operations, and how key LLM components such as attention, MLP, and Mixture-of-Experts (MoE) routing map onto the architecture..\n•\nDevelop and optimize kernels: Implement, test, and tune low-level kernels to improve the performance of LLM inference workloads..\n•\nAnalyze performance: Build analytical and roofline performance models, compare predictions against simulator and real-silicon measurements, and investigate performance bottlenecks..\n•\nBuild benchmarking and profiling tools: Develop benchmark harnesses, profiling scripts, and trace visualizations that make performance easier to measure, understand, and improve..\n•\nWork with AI coding agents: Explore workflows where AI agents assist with kernel development, optimization, and benchmarking, and identify the tools and context they need to work effectively..\n•\nCollaborate across engineering teams: Work closely with compiler, runtime, and kernel engineers to translate performance insights into practical improvements..\nExample project: Develop a benchmarking and profiling harness for an LLM building block, such as decode attention, at two abstraction levels. Compare measured performance against theoretical estimates and make the tooling accessible to AI coding agents.","requirements_text":"•\nCurrently enrolled in a Master's degree program in Computer Science, Electrical Engineering, Mathematics, or a related technical field..\n•\nStrong programming skills in Python and at least one of C, C++, or Rust..\n•\nUnderstanding of computer architecture, memory systems, bandwidth, and parallel computing through academic coursework or personal projects..\n•\nInterest in low-level programming, hardware performance, and optimization..\n•\nPractical experience using AI coding agents in development workflows, with the ability to evaluate and improve their output..\n•\nAn experimental mindset, with the ability to form hypotheses, analyze measurements, and communicate technical findings clearly..\n•\nCuriosity, initiative, and enthusiasm for learning new technologies..\nExperience in one or more of the following areas would be a plus:\n•\nGPU or AI accelerator programming..\n•\nEmbedded systems or bare-metal programming..\n•\nPerformance profiling, benchmarking, or optimization..\n•\nLLM inference fundamentals and architectures..\nWhy Join Tensordyne?\n•\nWork on cutting-edge AI technology: Contribute to the development of next-generation AI inference hardware and software..\n•\nGain hands-on experience: Apply your university knowledge to real silicon and challenging engineering problems..\n•\nOwn meaningful projects: Build tools and solutions that directly support our engineering teams..\n•\nLearn from experienced engineers: Work alongside specialists in hardware architecture, compilers, runtimes, and performance optimization..\n•\nExplore AI-assisted development: Gain practical experience building with AI coding agents and shaping how they interact with advanced hardware systems..\n•\nMake a lasting impact: Help develop performance tools and workflows that will support both current and future generations of Tensordyne technology..\nTensordyne's culture was built on the following values\n• Put people first. We only succeed when our people succeed.\n• Ethics and integrity always. Be open, honest, and respectful of everyone.\n• Think Big. Be ambitious and have audacious goals.\n• Aim for excellence. Quality and excellence count in everything we do.\n• Own it and get it done. Results matter!\n• Make each person better together than they would be as an individual.\n• Embrace each other’s differences, and embrace that there will be differences.\nTensordyne is an equal opportunity employer. We believe diverse teams are better equipped to solve complex problems and build exceptional technology. All qualified applicants will receive consideration for employment without regard to age, color, gender identity or expression, marital status, national origin, disability, protected veteran status, race, religion, pregnancy, sexual orientation, or any other characteristic protected by applicable laws, regulations, and ordinances.\nA note to recruitment agencies: Please do not contact Tensordyne employees or leaders regarding this role. We do not accept unsolicited agency resumes and are not responsible for fees associated with unsolicited submissions.","status":"open","duplicate_of":null,"first_seen":"2026-11-02T21:24:55.685306+00:00","responsibilities_summary":null,"requirements_summary":null,"all_tools":["Python","C","C++","Rust","Embedded Systems","LLM"],"posting_language":"en","evidence":[],"years_experience_max":null,"mobility_evidence":[],"model_version":"rules-8","english_badge":"English required","company_id":296}