← nvidia / Senior Technical Product Manager - GPU Direct Storage
tailored_resume_v2 / art_krhUp2L1Og4
role
model
anthropic/claude-sonnet-4.6
created
2026-05-20T22:42
↓ Download .docx ↓ Download .pdf PDF requires LibreOffice installed
What changed for nvidia
| change | why it matters |
|---|---|
| Summary rewritten to lead with GPU-accelerated AI workload benchmarking and distributed computing scale (675M+ engagements, 50K TPS) | JD's first hard requirement is technical expertise in accelerated computing and storage; NeurIPS credential and CUDA/MPS experience establish scientific computing credibility immediately |
| Intuit moved to lead the Experience section | Intuit has the strongest proof points for large-scale distributed platform PM, cloud-based infrastructure, and developer tooling — most directly mapped to JD's preferred qualifications |
| Intuit title unchanged; bullets reordered to lead with 675M+ / 50K TPS scale metric | JD emphasizes large-scale data management; enterprise throughput metrics are the strongest proof point and must appear in the first bullet |
| Splunk title reframed to 'Search Orchestration & Distributed Storage' (accurate — owned PostgreSQL storage service and Go microservices) | JD seeks distributed computing and storage solutions experience; reframe makes storage ownership explicit without inflating level |
| Splunk 10x performance improvement bullet reframed as 'HPC-environment tuning' analog | JD lists HPC environments as a hard requirement; query performance optimization at scale is the closest accurate mapping |
| StreamIO title reframed to 'GPU-Accelerated AI Platform' and RL Workbench bullet moved to lead | GPU Docker passthrough and CUDA/MPS benchmarking are the most NVIDIA-relevant StreamIO credentials; reframe makes GPU context explicit |
| RL Workbench moved to lead the Projects section | GPU Docker passthrough, CUDA/MPS training, and framework benchmarking (TRL, VeRL, NeMo RL) are the most directly relevant project credentials for an NVIDIA GPU-accelerated storage role |
| BRAIN project elevated to second position with NeurIPS and scientific computing framing | JD hard requirement: 'scientific computing, AI workloads, and HPC environments'; NeurIPS publication and C++ BPTT history establish deep technical credibility |
| Lawrence Berkeley National Laboratory added as standalone project entry | HPC laboratory research experience directly addresses JD's scientific computing requirement |
| Fintellect condensed to 1 bullet and moved to last in Experience | Lowest relevance to GPU storage/HPC role; kept for completeness and go-to-market credibility but deprioritized |
| Kaiser Permanente condensed to 2 bullets emphasizing 1.7 TB daily volume and Redis scalability | Large-scale data management and distributed storage infrastructure are the only Kaiser credentials relevant to this role |
| IBM condensed to 1 bullet | Software engineering background supports technical credibility claim but has minimal direct relevance to GPU storage PM role |
JD analysis (18 key phrases)
Key phrases: GPU-accelerated storageGPUDirect StoragecuFileaccelerated computinghigh-performance computingHPC environmentsAI workloadsscientific computingdistributed computinglarge-scale data managementgo-to-market strategycustomer engagementthought leadershiptechnical collateralproduct roadmapCUDA platformdata-intensive applicationsGPU programming
Hard requirements:
- BS/MS/PhD in CS, EE, Applied Math, or related technical discipline
- 12+ years relevant experience
- Strong technical expertise in accelerated computing and storage
- GPUDirect Storage, cuFile, or closely related technologies
- Scientific computing, AI workloads, HPC environments
- Exceptional interpersonal and communication skills
Preferred qualifications:
- Software development, architecture, or PM for distributed computing or storage solutions
- Direct experience with CUDA, GPU programming, large-scale data management
- Product strategy or requirements in cloud-based or HPC environments
Per-role mapping (10 roles scored)
| role | score | reframe angle | JD phrases that map |
|---|---|---|---|
| Streamio AI — Founder & CEO | 3/5 | GPU-aware compute platform and 0-to-1 product leadership; emphasize CUDA/MPS, data-intensive pipelines, and go-to-market | data-intensive applications, go-to-market strategy, AI workloads, accelerated computing |
| Fintellect AI — Founder & CEO | 2/5 | AI workloads and customer-driven product development; condense heavily | AI workloads, customer engagement |
| Intuit — Staff Product Manager | 4/5 | Large-scale distributed platform PM with developer tooling, cloud infrastructure, and data-intensive throughput at enterprise scale | large-scale data management, distributed computing, cloud-based environments, go-to-market strategy, developer engagement, product roadmap |
| Splunk — Senior Product Manager | 3/5 | Distributed search infrastructure and storage-layer PM; emphasize performance optimization and data-intensive query workloads | distributed computing, storage solutions, high-performance computing, data-intensive applications, product roadmap |
| Kaiser Permanente — SOA Technical PM | 2/5 | Large-scale data infrastructure and capacity planning; condense to 2 bullets | large-scale data management, distributed computing, storage solutions |
| IBM — Software Engineer | 2/5 | Software engineering background supporting technical credibility; 1 bullet | software development |
| Bank of America — Tech MBA Associate | 1/5 | Quantitative analytical foundation; 1 bullet | scientific computing |
| RL Workbench | 4/5 | GPU-accelerated AI workload benchmarking platform — most relevant project for NVIDIA HPC/AI context | GPU programming, AI workloads, accelerated computing, high-performance computing, data-intensive applications |
| aeval — AI Model Evaluation Platform | 3/5 | Data-intensive AI evaluation infrastructure with distributed storage stack | AI workloads, distributed computing, large-scale data management |
| BRAIN — Protein Structure Prediction ML Platform | 4/5 | Scientific computing and HPC-adjacent ML platform with NeurIPS publication — strongest academic/research credibility signal | scientific computing, AI workloads, high-performance computing, accelerated computing |
Tailored summary
Technical Product Manager with 12+ years driving developer-facing platforms and data-intensive infrastructure at enterprise scale — from hand-coding BPTT in C++ to benchmarking GPU-accelerated AI workloads across CUDA/MPS, TRL, VeRL, and NeMo RL today. Proven track record owning distributed computing and storage-layer products (675M+ engagements, 50K TPS) and delivering go-to-market strategy for developer tooling at Intuit and Splunk. NeurIPS-published researcher in scientific computing and neural architectures. BS in Computational Engineering Science, UC Berkeley; MS Software Management, Carnegie Mellon.