← cerebrassystems / Product Manager, Strategic Verticals
tailored_resume_v2 / art_fsf5gd17FbA
role
model
anthropic/claude-sonnet-4.6
created
2026-05-22T20:23
↓ Download .docx ↓ Download .pdf PDF requires LibreOffice installed
What changed for cerebrassystems
| change | why it matters |
|---|---|
| Summary rewritten to lead with 'technical product leader and former software engineer' framing | JD hard requirement is CS/EE or prior SWE background; leading with this satisfies the first filter |
| Summary embeds 'developer-facing platforms', '0-to-1', 'production-ready AI solutions', 'inference stack', and 'NeurIPS-published' | These are exact JD key phrases and preferred qualifications; embedding them naturally increases ATS and human reviewer resonance |
| OpenClaw multi-agent bullet reframed to explicitly connect to 'agentic computation use cases Cerebras' inference platform unlocks' | JD emphasizes agentic computation as a core Cerebras value prop; drawing the explicit line increases perceived fit |
| Fintellect AI condensed to 2 bullets focused on multi-provider LLM inference orchestration and inference stack tradeoffs | JD preferred qualification is 'experience with LLM serving stacks'; reframing Fintellect's multi-LLM routing as inference stack experience directly addresses this |
| Intuit lead bullet reordered to scale/latency metrics (675M engagements, sub-25ms TP99) rather than SDK work | JD values enterprise scale and latency performance; Cerebras' core pitch is speed, so mirroring latency metrics in the first Intuit bullet is highest-impact |
| Splunk 10x performance improvement bullet reframed to explicitly mirror Cerebras' '10x faster' positioning | Cerebras repeatedly uses '10x faster than GPU' as its core value prop; a candidate who has personally delivered 10x performance improvements resonates strongly |
| RL Workbench moved to lead the Projects section | JD preferred qualification explicitly mentions LLM serving stacks and fine-tuning; RL Workbench benchmarking TRL/VeRL/OpenRLHF/NeMo RL is the strongest proof of inference stack depth |
| RL Workbench first bullet adds 'directly maps to Cerebras' inference stack evaluation needs' | Makes the connection explicit for a hiring manager who may not immediately recognize the relevance of post-training RL to inference product work |
| Kaiser Permanente condensed to 1 bullet | Lower relevance score (2/5); space optimization for 2-page target while retaining enterprise platform credibility signal |
| IBM retained as single bullet | JD hard requirement includes 'prior SWE experience'; IBM satisfies this credential requirement and cannot be cut |
| Section order: Experience → Projects → Education → Teaching → Additional | JD values technical depth and hands-on AI work; Projects section (RL Workbench, aeval) provides critical technical credibility that should appear before Education |
| Streamio AI lead bullet reordered to OpenClaw multi-agent framework rather than Electron app details | JD explicitly calls out 'agent frameworks' as a preferred qualification; leading with OpenClaw immediately signals this fit |
JD analysis (20 key phrases)
Key phrases: 0-to-1 productsproduction-ready AI solutionsblazing-fast inferencewafer-scale architecturelighthouse accountsend-to-end customer journeymodel selection / fine-tuningbenchmark end-to-end performanceagentic computationdeveloper-facing productFortune 500 enterprisesAI-native startupsstrategic verticalsco-architect solutionsdrive the product roadmapGTM strategistreal-time serving speedssub-ms inference latenciessovereign AIT-shaped
Hard requirements:
- Strong CS/EE background or prior SWE experience
- 5+ years PM or SWE experience at Senior+ level
- Familiarity with LLMs, inference, agents
- Customer-facing communication in complex, high-stakes scenarios
- 0-to-1 product ownership end-to-end
- Cross-functional stakeholder alignment (Sales, SA, Engineering, Product)
- GTM strategy and execution
Preferred qualifications:
- Experience with LLM serving stacks (vLLM, TensorRT-LLM, TGI)
- Agent frameworks experience
- Interest in developer platforms and tooling
- MBA or equivalent
Per-role mapping (10 roles scored)
| role | score | reframe angle | JD phrases that map |
|---|---|---|---|
| Streamio AI — Founder & CEO | 4/5 | 0-to-1 AI platform founder with hands-on LLM inference and multi-agent orchestration — directly mirrors Cerebras' inference + agentic computation narrative | 0-to-1 products, production-ready AI solutions, agentic computation, agent frameworks, end-to-end customer journey, AI-native startups |
| Fintellect AI — Founder & CEO | 3/5 | Vertical AI product (fintech) with multi-LLM inference orchestration — demonstrates strategic verticals instinct and inference stack familiarity | strategic verticals, model selection, agentic computation, AI-native startups, production-ready AI solutions |
| Intuit — Staff PM, Developer Frameworks & Platform Infrastructure | 5/5 | Enterprise-scale developer platform PM with proven latency optimization and SDK tooling — maps directly to Cerebras' developer-facing inference platform and Fortune 500 customer base | developer-facing product, Fortune 500 enterprises, benchmark end-to-end performance, drive the product roadmap, co-architect solutions, production-ready AI solutions, real-time serving speeds |
| Splunk — Senior PM, Search Orchestration | 3/5 | Technical PM owning microservices and query performance — 10x improvement narrative resonates with Cerebras' 10x inference speed positioning | benchmark end-to-end performance, Fortune 500 enterprises, drive the product roadmap |
| Kaiser Permanente — SOA Technical PM | 2/5 | Enterprise platform PM with infrastructure scale — condense to 1-2 bullets | Fortune 500 enterprises, strategic verticals |
| IBM — Software Engineer | 2/5 | SWE foundation — keep minimal to satisfy 'prior SWE experience' hard requirement | — |
| Bank of America Merrill Lynch — Tech MBA Associate | 1/5 | Condense to single bullet; signals MBA + finance vertical depth relevant to enterprise customer base | — |
| RL Workbench | 5/5 | Lead project — directly demonstrates LLM training/inference stack fluency Cerebras values | benchmark end-to-end performance, model selection / fine-tuning, LLMs, inference |
| aeval — AI Model Evaluation Platform | 4/5 | Inference quality evaluation tooling — maps to Cerebras' need to benchmark and validate model outputs for customers | benchmark end-to-end performance, production-ready AI solutions, model selection |
| BRAIN — Protein Structure Prediction | 3/5 | NeurIPS research + 8B param model — signals serious ML depth beyond typical PM | LLMs, publish and open source cutting-edge AI research |
Tailored summary
Technical product leader and former software engineer with 12+ years building developer-facing platforms and AI products at scale — from shipping production-ready AI solutions as a 0-to-1 founder to scaling a developer inference platform to 675M+ engagements and sub-25ms TP99 at Intuit. Hands-on LLM inference and multi-agent orchestration experience (OpenClaw, RAG pipelines, multi-provider LLM routing) with deep familiarity across the inference stack. NeurIPS-published ML researcher; benchmarked GRPO/DPO across TRL, VeRL, OpenRLHF, and NeMo RL. CMU MBA.