← cursor / Product Manager, Cloud Agents
tailored_resume_v2 / art_ds69WLnoOzc
role
model
anthropic/claude-sonnet-4.6
created
2026-05-24T20:50
↓ Download .docx ↓ Download .pdf PDF requires LibreOffice installed
What changed for cursor
| change | why it matters |
|---|---|
| Summary rewritten to lead with 'developer tools, infrastructure platforms, and AI-powered applications at scale' and explicitly name agent orchestration, artifact review pipelines, and sandboxed environments | JD's first requirement is shipping developer tools/infrastructure/AI-powered apps as a PM; these are Cursor's exact words |
| Streamio AI bullets reordered to lead with OpenClaw multi-agent orchestration framework | JD's core product is agent orchestration; OpenClaw is the strongest direct proof point on the resume |
| Streamio AI HLS pipeline bullet reframed as 'artifact and review layer (video recordings, live previews)' | JD explicitly calls out logs, video recordings, and live previews as the agent output surface — this is a genuine match reframed in JD language |
| Intuit ICE Self-Service bullet reframed to explicitly connect developer onboarding time-to-value to cloud agent task handoff experience | JD asks PM to own the task handoff experience; reducing onboarding from weeks to minutes is the strongest time-to-value proof point |
| Splunk Scheduler Service bullet reframed as 'async task execution, task queuing, retries, and error recovery' analogue | JD explicitly lists retries, error recovery, and parallelism as core orchestration responsibilities; Scheduler Service is a genuine match |
| Kaiser Permanente Logging-as-a-Service reframed as 'observability and artifact layer for self-hosted enterprise deployment' | JD calls out self-hosted/enterprise deployment model as a key project area; 1.7TB/day LaaS is genuine enterprise infrastructure PM experience |
| RL Workbench moved to lead the Projects section | GPU Docker passthrough + sandboxed framework benchmarking is the strongest technical proof of sandboxing and resource allocation intuition — JD's core orchestration concerns |
| AutoEval moved to second project position | Directly maps to JD's artifact and review layer requirement — video/screenshot/structured output review reducing evaluation from 72h to 4min is a compelling proof point |
| RL Workbench bullets reframed to explicitly call out sandboxing, resource allocation, and parallelism language | JD uses these exact terms for the orchestration model the PM will own |
| AutoEval bullets reframed to mirror JD's 'logs, screenshots, video recordings, and live previews' language | Genuine match — AutoEval does exactly this for robot model outputs; JD language applied accurately |
| aeval bullets reframed to emphasize 'measurement layer' and 'task completion rate' instrumentation | JD explicitly asks PM to instrument success: task completion rate, time-to-value, cost-per-task, developer trust |
| Fintellect AI condensed to 2 bullets emphasizing multi-provider LLM orchestration and fleet-level reliability/cost tradeoffs | Role is relevant but secondary to Streamio; condensed to preserve space while keeping the LLM orchestration and cost-per-task signal |
| IBM and BofA each condensed to 1 bullet | Low relevance to Cloud Agents PM role; kept for career completeness per anti-pattern rules but space-optimized |
| Deep Learning Education Platform project removed from projects section | Lowest relevance to Cloud Agents role; space optimization for 2-page target while keeping all 4 remaining projects which have direct JD mapping |
JD analysis (20 key phrases)
Key phrases: cloud agentsagent orchestrationtask handoff experiencesandboxingparallelism and retriesartifact and review layerlogs, screenshots, video recordings, and live previewsdeveloper trustself-hosted and enterprise deploymentfleet of agentsprovisioned, orchestrated, and monitoredfast, trustworthy, and worth relying ontask completion ratetime-to-valuecost-per-taskdeveloper toolsinfrastructure productsAI-powered applicationsdistributed systemsdeveloper experience
Hard requirements:
- Shipped developer tools, infrastructure products, or AI-powered applications as a PM
- Understands distributed systems (orchestration, sandboxing, reliability)
- Strong product intuition for developer experience
- Deeply technical — can read PRs, reason about architecture, form tradeoff opinions
- Comfortable making decisions with incomplete information
- Clear written and verbal communication
- Uses AI coding tools regularly with strong opinions
Preferred qualifications:
- Experience with agent orchestration models
- Experience with artifact/review layers (logs, screenshots, video, live previews)
- Experience with enterprise/self-hosted deployment models
- Experience instrumenting success metrics (task completion rate, time-to-value, cost-per-task)
- Experience designing task handoff UX
- Experience with sandboxing, parallelism, retries, error recovery, resource allocation
Per-role mapping (10 roles scored)
| role | score | reframe angle | JD phrases that map |
|---|---|---|---|
| Streamio AI — Founder & CEO | 5/5 | Lead with multi-agent orchestration and artifact/review pipeline as direct analogues to Cloud Agents product surface | agent orchestration, logs, screenshots, video recordings, and live previews, provisioned, orchestrated, and monitored, AI-powered applications, developer tools, task handoff experience |
| Fintellect AI — Founder & CEO | 3/5 | Emphasize multi-agent orchestration and LLM routing as distributed AI systems experience | AI-powered applications, agent orchestration, fleet of agents |
| Intuit — Staff PM | 5/5 | Frame as owning developer-facing infrastructure platform at massive scale — direct parallel to cloud agent provisioning and orchestration | provisioned, orchestrated, and monitored, developer tools, infrastructure products, developer experience, distributed systems, self-hosted and enterprise deployment, task completion rate, time-to-value |
| Splunk — Senior PM | 4/5 | Frame Search Orchestration as distributed systems PM experience with async task scheduling — maps to cloud agent task queuing and orchestration | orchestration, distributed systems, sandboxing, parallelism and retries, infrastructure products |
| Kaiser Permanente — SOA Technical PM | 3/5 | Frame as enterprise infrastructure PM with observability/logging platform ownership | self-hosted and enterprise deployment, infrastructure products, distributed systems |
| IBM — Software Engineer | 2/5 | Condense to single bullet showing technical depth and enterprise credibility | — |
| Bank of America — Tech MBA Associate | 1/5 | Condense to single bullet or merge; keep for credential completeness | — |
| RL Workbench | 4/5 | Lead projects section — GPU Docker passthrough and framework benchmarking maps directly to cloud agent sandboxing and orchestration | sandboxing, parallelism, cost-per-task, provisioned, orchestrated, and monitored |
| aeval — AI Model Evaluation Platform | 4/5 | Frame as instrumenting AI system success — maps to task completion rate, developer trust, cost-per-task measurement layer | task completion rate, developer trust, instrumenting success, artifact and review layer |
| AutoEval | 4/5 | Frame as artifact and review layer for AI agents — logs, video, structured output review | logs, screenshots, video recordings, and live previews, artifact and review layer, developer trust, time-to-value |
Tailored summary
Technical PM with 12+ years shipping developer tools, infrastructure platforms, and AI-powered applications at scale — from owning agent orchestration frameworks and artifact review pipelines at Streamio AI to scaling platform infrastructure to 675M+ engagements at Intuit. Built multi-agent orchestration (OpenClaw), real-time video/preview artifact layers, and sandboxed RL benchmarking environments with GPU Docker passthrough. Deep distributed systems fluency: microservice orchestration at Splunk, 50K TPS rSocket migrations, and hands-on RL post-training workbenches. NeurIPS published; UC Berkeley CS + CMU MBA.