Christi Akinwumi
Conversation Designer and AI Product Manager
Dallas-Fort Worth, TX ·christi.io ·[email protected] ·linkedin.com/in/christiakinwumi ·github.com/mrskinzo ·huggingface.co/christiakinwumi
Role lens
Conversation Design
Conversation Designer and AI Product Manager · Intent, Context, Verify · The translation layer for AI teams · Plain-language advocate
The translation layer is my role. I sit between people who speak past each other. Intent, Context, Verify is my method. Intent sorts what people need. Context gives the system the right knowledge. Verify tests the answer before customers see it.
Use this lens when the question is language, boundaries, trust, and failure handling.
- Rebuilt a routing taxonomy from 50K real conversations into mutually exclusive intents that reduced misrouting 30%.
- Designed self-service and escalation flows with localization, decision trees, and clearer customer language.
- Built live tool-using flows that separate tentative states from confirmed actions.
AI Product Management
AI Product Manager and Conversation Designer · Thoughtful Execution · Intent, Context, Verify · Plain-language advocate
Thoughtful Execution is how I run a project. I identify the friction points. I test approaches on real work. I keep what moves the metric.
Use this lens when the question is product strategy, rollout, and measurable value.
- Led product strategy and conversation design for AI support experiences serving 20M+ users across 8+ markets.
- Defined requirements, evaluation criteria, and launch decisions for routing, retrieval, and model changes.
- Turned customer and operational evidence into roadmap priorities, experiments, and investment recommendations.
Highlights
- Led AI product strategy and conversation design for experiences serving 20M+ users across 8+ international markets.
- Led product definition for LLM routing and grounded retrieval, cutting median response time 25% and grounding answers in verified support content.
- Built and operate a live booking agent that connects the Anthropic API, Microsoft Graph, and Twilio for a DFW service business.
Experience
Founder & Principal Consultant · Intelligent CX Consulting, LLC
Dallas-Fort Worth, TX (remote, US clients)
- Serve as embedded conversational AI consultant to an enterprise client, leading conversation design, prompt design, and UX strategy for production AI experiences (ongoing since April 2026).
- Run scoped conversational AI engagements focused on agent workflows, evaluation, intent taxonomies, and support architecture.
- Designed and built Boothly, a tool-using Claude agent for a photo booth business. A completed website flow produces a sourced quote, a tentative Outlook hold, and an owner SMS in under 60 seconds.
- Defined the product requirements, conversation flows, tool contracts, fallback behavior, and operating guardrails, then implemented the production system.
Conversation Designer · GoDaddy
- Led product strategy and conversation design for AI support experiences serving 20M+ users across 8+ markets, working across product, engineering, data science, support, and operations.
- Led product definition for an LLM routing system that cut median response time 25%.
- Defined the evaluation question, golden-set requirements, decision criteria, and recommendation for a six-week RAG comparison. The selected hybrid retrieval architecture later improved answer accuracy 45% against the pre-RAG baseline.
- Led an intent taxonomy redesign based on 50K conversations, restructuring intents to be mutually exclusive and reducing misrouting 30%.
- Contributed product, CX, and evaluation criteria to a comparison of OpenAI, Anthropic, and AWS Bedrock. Co-developed ROI recommendations that supported $500K+ in investment.
Workflow Author · Peloton
- Rebuilt self-service for a connected fitness product across 8 international markets: information architecture, decision trees, localization, and a product gap feedback loop that fed the roadmap.
- Resolution time down 40%, self-service adoption up 35%, support volume down 20%.
Selected projects
- Boothly Booking Concierge. Live tool-use agent for a real business: Claude (Anthropic SDK), Microsoft Graph, Twilio. Case study at christi.io/work/boothly-booking-concierge.
- RAG Pipeline Comparison. Public Python evaluation harness comparing three configurations across chunk size, overlap, cosine retrieval, and MMR. github.com/mrskinzo/rag-pipeline-comparison.
- christi.io. This site: Astro, a custom build pipeline, and a Claude powered chatbot grounded in my published writing.
Education
- M.S., Curriculum & Instruction
Texas A&M International University - Post-Graduate Certification, Technical Communication
University of North Texas - B.G.S., Business
Texas Woman's University - MA, Technical Communications (expected May 2027)
In progress. Not yet earned.
Certifications
Skills
- Conversation design
- Multi-agent orchestration
- Prompt architecture
- RAG evaluation
- Eval harness design
- Intent taxonomy
- LangFuse
- Dialogflow CX
- Azure AI Foundry
- AWS Bedrock
- Claude (Anthropic)
- JavaScript
- Python