← amplitude / Principal Product Manager, AI Agents & MCP
tailored_resume_v2 / art_jnxCLQnFlCM
role
model
anthropic/claude-sonnet-4.6
created
2026-09-08T21:11
↓ Download .docx ↓ Download .pdf PDF requires LibreOffice installed
What changed for amplitude
| change | why it matters |
|---|---|
| Summary rewritten to lead with MCP server delivery and multi-agent orchestration | JD's primary requirement is hands-on agentic product delivery including MCP surface ownership |
| Summary embeds 'evals', 'probabilistic model quality', 'MCP servers', 'coding agents', '0-to-1', and '675M+ engagements' | JD key phrases: evals, MCP, agentic products, 0-to-1, feedback mechanisms, probabilistic product |
| Streamio AI bullets reordered to lead with MCP server delivery | MCP is the most direct proof point for Amplitude's MCP surface ownership requirement |
| OpenClaw multi-agent orchestration elevated to second bullet in Streamio | Agent orchestration framework directly maps to 'AI agent strategy' and 'agentic products' requirements |
| Vantage automated AI judge framed as 'quality standards and feedback mechanisms for a probabilistic AI product' | JD requires experience building systems that measure whether a probabilistic product is getting better |
| Fintellect LLM orchestration bullet reframed to emphasize 'context management and latency engineering' | JD explicitly requires fluency in context management and latency as LLM competencies |
| RL Workbench moved to lead the projects section | Strongest proof of evals, quality standards, and measuring probabilistic model improvement — top JD requirement |
| aeval project framed around 'feedback mechanism for measuring whether a probabilistic AI product is improving over time' | Direct mirror of JD language: 'experience building systems that measure whether a probabilistic product is getting better over time' |
| Intuit section condensed from 8 to 4 bullets, leading with 675M+ scale metric | Scale credibility matters for Amplitude's enterprise context; developer tooling bullets support platform PM credibility |
| Kaiser Permanente condensed to 1 bullet | Low relevance to agentic AI role; space optimization for higher-relevance content |
| IBM and BofA each kept at 1 bullet | Career completeness; space constraints; low direct relevance to agentic AI PM role |
| Deep Learning Education Platform project removed | Lowest relevance to agentic AI PM role; space optimization |
JD analysis (20 key phrases)
Key phrases: AI agentsMCPagentic products0-to-1evalsquality standardsfeedback mechanismscontext managementprobabilistic productrapid iterationthought leaderagent strategycustomer workflowsLLM trendscross-functional partnershipbias toward rapid iterationget to insights fasterMCP capabilitiescoding agentagent experience
Hard requirements:
- 7+ years product management experience
- 0-to-1 product delivery
- 3+ years building AI/ML products
- Hands-on agentic product delivery
- LLM fluency: context management, latency, eval methods, failure modes
- Experience measuring probabilistic product quality over time
- Cross-functional partnership with engineering and design
- Comfort with ambiguity and rapidly evolving roadmaps
Preferred qualifications:
- MCP surface ownership
- Thought leadership on agentic products
- Field enablement and go-to-market partnership
- Customer discovery and workflow analysis
Per-role mapping (10 roles scored)
| role | score | reframe angle | JD phrases that map |
|---|---|---|---|
| Streamio AI — Founder & CEO (Vantage + StreamIO) | 5/5 | Lead with MCP server delivery and multi-agent orchestration as direct proof of agentic product ownership; emphasize 0-to-1 execution and customer discovery | MCP, AI agents, agentic products, 0-to-1, coding agent, agent experience, customer workflows, rapid iteration |
| Fintellect AI — Founder & CEO | 4/5 | Frame as multi-agent LLM orchestration with context management and structured-output validation — maps to LLM fluency and probabilistic product quality requirements | AI agents, LLM trends, context management, agentic products, 0-to-1, feedback mechanisms |
| Intuit — Staff PM, Developer Frameworks & Platform Infrastructure | 3/5 | Frame as enterprise-scale platform product leadership with developer tooling and data-driven prioritization — supports credibility at Amplitude's scale | cross-functional partnership, quality standards, feedback mechanisms, rapid iteration |
| Splunk — Senior PM, Search Orchestration | 3/5 | Frame as orchestration and query-performance product leadership with rapid 0-to-1 delivery | rapid iteration, 0-to-1, cross-functional partnership |
| Kaiser Permanente — SOA Technical PM | 2/5 | Condense to 1 bullet emphasizing enterprise platform scale | — |
| IBM — Software Engineer | 1/5 | Single bullet, keep for career completeness | — |
| Bank of America Merrill Lynch — Tech MBA Associate | 1/5 | Single bullet, keep for career completeness | — |
| RL Workbench (Project) | 5/5 | Lead projects section — strongest proof of evals, quality standards, and measuring probabilistic product improvement over time | evals, quality standards, feedback mechanisms, probabilistic product, LLM trends |
| aeval — AI Model Evaluation Platform (Project) | 5/5 | Second project — reinforces evals and measuring probabilistic product quality over time | evals, quality standards, feedback mechanisms, probabilistic product, agent experience |
| AutoEval / BRAIN / Deep Learning Platform (Projects) | 3/5 | Condense; keep AutoEval for eval proof, BRAIN for NeurIPS credibility | evals, LLM trends |
Tailored summary
Principal PM and hands-on AI builder with 12+ years delivering 0-to-1 products — including production MCP servers, multi-agent orchestration frameworks, and LLM eval platforms — at the intersection of agentic AI and developer tooling. Shipped MCP SDK integrations exposing AI tools to coding agents, built OpenClaw multi-agent gateway with subagent delegation across four industry verticals, and engineered aeval: a statistical eval platform measuring probabilistic model quality over time with automated safety gates. Scaled AI platforms to 675M+ engagements at Intuit; NeurIPS published researcher.