{"id":2093130,"url":"https://alion.io/job/hedra-inference-optimization-engineer","title":"Inference Optimization Engineer","company":{"id":4619,"name":"Hedra","domain":"hedra.com","url":"https://alion.io/company/hedra","size_band":"1001-5000","is_staffing_agency":false,"employer_type":"direct","is_intermediary":false,"listed_via":null,"ats_vendor":"Ashby","truth_index":{"grade":"B","score":75,"open_postings":3,"ghost_share":0,"stale_share":1,"repost_share":0,"time_to_fill_p50_days":null,"computed_at":"2026-10-10T05:45:15Z"}},"role":"AI/ML","role_family":"AI/ML","seniority":null,"employment_type":"full_time","work_mode":"on_site","remote_scope":null,"remote_scope_basis":null,"remote_working_hours":null,"hiring_geo_confidence":"structured","locations":["San Francisco, United States"],"countries":["US"],"hiring_countries":[],"hiring_countries_total":0,"salary":null,"salary_estimate":{"min_usd":166000,"max_usd":366000,"period":"year","method":"role_country_seniority_unknown","sample_n":3798},"experience_years_min":null,"visa_sponsorship":false,"relocation_package":false,"has_equity":false,"technologies":[{"name":"C++","optional":false},{"name":"CUDA","optional":false},{"name":"CUDA Toolkit","optional":false},{"name":"Python","optional":false},{"name":"PyTorch","optional":false},{"name":"PyTorch C++","optional":false},{"name":"Quantization","optional":false},{"name":"SGLang","optional":false},{"name":"TensorRT","optional":false},{"name":"Triton","optional":false},{"name":"vLLM","optional":false},{"name":"CUTLASS","optional":true},{"name":"Multimodal AI","optional":true}],"status":"live","first_seen_at":"2026-10-08T14:26:27Z","employer_posted_date":"2026-10-08","last_verified_at":"2026-10-12T00:21:19Z","board_verified":true,"closed_at":null,"days_open":3,"trust":{"level":"ok","repost_count":null,"flags":[],"days_open":3},"description":"About Hedra\nHedra is the platform, models, and infrastructure for visual intelligence.\nWe build models and systems that push the frontier of visual intelligence, along with the infrastructure required to make those models fast, efficient, reliable, and accessible at scale.\nWe’re a small, highly technical team in San Francisco, backed by a16z and other leading investors. Researchers and engineers at Hedra work closely across boundaries, own problems end to end, and have significant influence over both what we build and how we build it.\nThe Role\nWe’re looking for an Inference Optimization Engineer to work alongside our research team on making state-of-the-art visual models fast and efficient at inference time.\nYou’ll work at the boundary between research and systems, taking new model architectures and figuring out how to run them efficiently on modern hardware. That means understanding where time and memory are being spent, identifying opportunities for algorithmic and systems-level improvements, and implementing optimizations across model architecture, inference algorithms, runtimes, kernels, and distributed execution.\nThe problems rarely live neatly within one layer of the stack. Depending on what you find, you might modify how a model executes, develop a new inference technique, write a custom GPU kernel, rethink memory movement, or change how work is distributed across accelerators.\nWe care more about technical depth, curiosity, and demonstrated ability than years of experience. We’re open to experienced ML systems engineers as well as exceptional early-career engineers or researchers who have already gone unusually deep on model performance, GPU systems, or efficient inference.\nWhat You’ll Do\nWork directly with research scientists and engineers to make new visual models fast and efficient at inference time.\n\nProfile model architectures and workloads to understand bottlenecks across compute, memory, communication, and model execution.\n\nDevelop and implement new approaches to improving inference latency, throughput, memory efficiency, and GPU utilization.\n\nExplore algorithmic optimizations including quantization, sparsity, caching, compilation, attention optimizations, and alternative execution strategies.\n\nBuild or optimize GPU kernels using CUDA, Triton, or similar technologies when existing implementations leave performance on the table.\n\nOptimize model execution across single-GPU, multi-GPU, and multi-node environments.\n\nReason about the interaction between model architecture and hardware, and work with researchers when architectural changes can unlock meaningful performance improvements.\n\nInvestigate communication, memory movement, parallelism, and distributed execution strategies for large visual models.\n\nBuild rigorous benchmarking, profiling, and performance-regression infrastructure to understand performance and evaluate new optimization ideas.\n\nEvaluate new inference runtimes, compilers, frameworks, optimization techniques, and accelerator hardware.\n\nStay close to advances in efficient inference, GPU programming, model architectures, compilers, and ML systems research, and rapidly test promising ideas.\n\nHelp turn research breakthroughs into models that can be deployed and served efficiently at scale.\n\nWhat We’re Looking For\nDeep technical ability in efficient ML inference, ML systems, GPU computing, or adjacent research, demonstrated through research, production engineering, open-source contributions, or unusually ambitious independent work.\n\nStrong understanding of how modern deep learning models execute on hardware, including the relationship between compute, memory, communication, and performance.\n\nExperience profiling ML workloads, identifying bottlenecks, forming hypotheses, and driving measurable performance improvements.\n\nStrong programming fundamentals in Python, C++, or another systems-oriented language.\n\nExperience with some combination of PyTorch, CUDA, Triton, TensorRT, vLLM, SGLang, or comparable technologies.\n\nAbility to reason across abstraction layers rather than treating model architecture, framework, runtime, kernel, and hardware boundaries as fixed.\n\nStrong intuition for performance tradeoffs across latency, throughput, memory, numerical precision, model quality, and complexity.\n\nCuriosity about how models work internally and a willingness to modify or rethink existing approaches when the performance problem calls for it.\n\nComfort working on ambiguous problems where the bottleneck, and sometimes even the right question, is not known in advance.\n\nAbility to communicate technical ideas clearly and collaborate closely with research scientists and engineers.\n\nWe don’t expect every candidate to have experience across the entire stack. Exceptional depth in one or more relevant areas, combined with the ability and curiosity to reason across the others, matters more to us than checking every box.\nNice to Have\nExperience optimizing large generative, multimodal, vision, or video models.\n\nCUDA, Triton, CUTLASS, or other GPU kernel development.\n\nDeep knowledge of GPU architecture, memory hierarchy, and hardware-aware optimization.\n\nExperience with attention optimization, kernel fusion, memory-efficient execution, or custom operators.\n\nModel compilation or graph optimization experience.\n\nQuantization, sparsity, caching, speculative execution, or other efficient inference techniques.\n\nExperience optimizing diffusion, autoregressive, transformer, or other large generative architectures.\n\nMulti-GPU or multi-node model execution, including tensor, pipeline, sequence, or other forms of parallelism.\n\nExperience optimizing communication or data movement between accelerators.\n\nExperience with profiling tools such as Nsight Systems or Nsight Compute.\n\nContributions to ML systems, inference runtimes, compilers, GPU libraries, or performance-focused open-source projects.\n\nResearch or publications in efficient ML, ML systems, GPU computing, compilers, or related areas.\n\nBenefits:\nCompetitive compensation and equity\n\n401k\n\nHealthcare (Silver PPO Medical, Vision, Dental)\n\nLunch and snacks at the office\n\nThis role is based in San Francisco, and we work together in person five days a week.","description_format":"text","description_chars":6207,"description_truncated":false,"requirements":{"experience_years_min":null,"management_years_min":null,"team_size_min":null,"manages_managers":false,"education":null,"security_clearance":false,"languages":[]},"benefits":["401k plan","Equity"],"hiring_locations":[],"hiring_excludes":[],"relocation_offered":false,"industries":["Generative Image","Generative Video","Generative Audio & Music"],"lifecycle":[{"event":"open","at":"2026-10-08T15:46:00Z"}],"visa":[],"liveness":{"score":86,"band":"hot","label":"Hiring now","p_open":1,"p_active":0.86,"p_room":1,"age_days":1,"expected_fill_days":20,"reasons":["conf:2","win:early"],"computed_at":"2026-10-10T05:45:15Z"},"pay":null,"html_url":"https://alion.io/job/hedra-inference-optimization-engineer","json_url":"https://alion.io/job/hedra-inference-optimization-engineer.json","meta":{"generated_at":"2026-10-12T00:39:11Z","cache_seconds":300,"methodology":"https://alion.io/methodology","terms":"https://alion.io/terms","contact":"https://alion.io/contact","api":"https://alion.io/developers","about":"Alion is a live layer of people, companies and AI agents: who they are, whether they are real and active right now, what they do and how to work with them, readable by people and by agents and paid per call.","catalog":"https://alion.io/catalog.json","usage":{"tier":"crawler","counted_by":"address","units_charged":1,"used_today":10,"day_limit":5000,"remaining_today":4990,"minute_limit":60,"resets_at":"2026-10-13T00:00:00Z"}},"offers":[{"id":"company.slices","title":"One company in depth, by slice","status":"live","price":{"credits":0.02,"usd":0.002,"plus_per_slice":{"credits":0.05,"usd":0.005}},"unit":"per company, plus each slice with data","note":"the employer in depth","call":{"mcp_tool":"get_company","arguments":{"id":4619},"rest":"https://alion.io/mcp/rest/get_company?id=4619"},"human":"https://alion.io/catalog?offer=company.slices&for=job%2Fhedra-inference-optimization-engineer"},{"id":"market.stats","title":"A market slice: pay, demand and time to fill","status":"live","price":{"credits":1,"usd":0.1},"unit":"per slice","note":"pay, demand and time to fill for this role and place","call":{"mcp_tool":"market_stats"},"human":"https://alion.io/catalog?offer=market.stats&for=job%2Fhedra-inference-optimization-engineer"},{"id":"job.search","title":"Open jobs by role, technology, place, pay and visa","status":"live","price":{"credits":0.02,"usd":0.002},"unit":"per posting in a list","note":"similar open postings","call":{"mcp_tool":"search_jobs"},"human":"https://alion.io/catalog?offer=job.search&for=job%2Fhedra-inference-optimization-engineer"},{"id":"company.verify","title":"Is this company real and active right now","status":"pilot","price":null,"unit":"per company","request":{"url":"https://alion.io/catalog/request","method":"POST","body":"{\"offer\": \"company.verify\", \"for\": \"job/hedra-inference-optimization-engineer\", \"note\": \"what you need it for\"}"},"human":"https://alion.io/catalog?offer=company.verify&for=job%2Fhedra-inference-optimization-engineer"}]}