01 / 04
$250M
ARR org-wide — the revenue context my scoring and assignment systems operate in.
John KipeAI Systems Architect
I build AI agentspeople actually use.
Most of the work starts messy: scattered customer data, manual decisions, and a process held together by people. I turn that into agents and tools revenue teams can rely on.
Built with real data, human review, and a clear owner after launch.
04 receipts
The operating context behind the work.
02 / 04
13+
AI agents shipped to production — deployed and operating, not demos.
03 / 04
1,200+
accounts scored and prioritized on a standing 90-day cadence.
04 / 04
15
person team led and developed while these systems went live.
01
Selected work
Three systems, each with an outcome you can inspect.
The problem, implementation, and operational result stay together so the work can be evaluated on more than polish.
Featured evidence
01 / 03Transition Readiness Agent
CSMs inherited accounts with no compiled context after onboarding handoff.
What I built
Built a Glean Agent synthesizing 8+ systems — Salesforce, Gong, Totango, Provisio, Zendesk, Gmail, Drive, Slack — into a ready-made transition package with CSM talking points.
Operational result
CSMs get full customer intelligence compiled automatically, no manual reconstruction needed.
- Phase
- Discover → Deliver
- Scale
- 1,200+ accounts
- Result
- One agent, full picture
Glean Agent / SOQL / Salesforce / Gong / Totango / Provisio
02Case study
REACH Scoring Agent
Account prioritization depended on gut feel — no standardized scoring across 1,200+ accounts.
Built: Built a Glean Agent scoring accounts on a 90-day cadence and key events, with Slack notifications, human review, and one-click Totango import with historical tracking.
Glean Agent / Salesforce / Totango / Provisio / Gong / Slack / Qualtrics
03Case study
CSM Assignment Engine
Assigning accounts purely on headcount count creates hidden imbalance — a $50K single-product account and a $500K five-product account are treated identically, overloading CSMs with lower headcount but higher complexity.
Built: Designed a Constrained Greedy Heuristic that evaluates four dimensions (Volume 35%, ARR 25%, Renewal Balance 25%, Portfolio Risk 15%) with dynamic capacity re-calculation. Includes the Bouncer pattern for white-glove Enterprise handling, ARR anchoring (vs TCV) to avoid renewal penalties, and a New Hire Magnet for accelerated ramp.
Google Apps Script / Totango API / Salesforce API / Glean API / Constrained Greedy Heuristic / Bouncer Pattern
02
Capabilities
One delivery sequence, from context to handoff.
Discovery through deployment stays connected: constraints shape the architecture, governance shapes the build, and observation shapes what happens next.
01Discover
Surface real constraints
Stakeholder interviews, system audits, and data inventory — identifying what's actually broken before proposing architecture.
02Architect
Design governed flows
Information architecture, workflow design, and quality gates — systems that enforce constraints without blocking progress.
03Dispatch
Route with precision
Task distribution, load balancing, and failure handling — getting work to the right system at the right time.
04Govern
Enforce policy gates
Compliance checks, validation layers, and audit trails — building trust through transparency, not friction.
05Observe
Measure what matters
Telemetry, success metrics, and feedback loops — knowing what's working because you're watching, not guessing.
06Ship
Deploy with confidence
Rollout strategies, rollback paths, and operational handoff — shipping is the start, not the finish line.
03
Next step
Proof of work, built to inspect.
Case studies, operating models, and delivery context behind the résumé—organized for people evaluating how I think and build.