Katy Simmons

Selected work

Problems worth getting obsessed with.

Not every project deserves a case study. These do. Each one runs the same spine, so you can compare how I think across very different problems.

Featured work

Start here
012019–2024

Scaling Apptegy from 800 to 4,000 Customers

Five years taking an EdTech portfolio from startup to scale, 5X customer growth and a contribution to $20M ARR, built on operating models that outlasted my tenure.

800 → 4,000

Customers

ApptegyEdTech
Read the case study
022024–2026

MAIA: AI-Generated Content Detection at Platform Scale

MAIA detects AI-generated audio at roughly 99.36% accuracy and 99.75% precision across roughly 700K tracks a day, with 5 to 7 second latency. The model was the easy half. The work was deciding what tagging, policy, review, monetization, and artist impact looked like on the other side of a detection.

99.36%

Detection accuracy

SoundCloudMusic & Media
Read the case study
032026–Present

Rebuilding Finance Close with AI

Finance spent a full week every month manually recalculating ARR and reconciling it against contract lines. I personally built the first version end to end, and 89% of contracts reconciled automatically end to end. The remaining 11% went to people on purpose.

89%

of contracts reconciled automatically end to end

HoneycombDeveloper Tools
Read the case study
042025–2026

Turning Consumer Support Into a Product

Consumer support was spread across The Weather Channel, Weather Underground, and Storm Radar, across platforms, channels, and knowledge sources, measured by throughput. I rebuilt it as one product: AI entry, human support, knowledge, routing, taxonomy, analytics, and a feedback path into Product. The AI matters here, but the system is the point.

3 brands

One support product across all of them

The Weather ChannelWeather & Data
Read the case study
Browse

12 case studies

012026

AI Transformation: The Evidence Changed the Question

I ran 36 discovery sessions to find where AI could help. The evidence said model capability was almost never the blocker. Context, ownership, connected systems, and trust were. So I changed the question and built the method around it.

36

Discovery sessions across functions

HoneycombDeveloper Tools
Read the case study
022026

GTM Operating System: Radical Frictionlessness

Isolated AI wins do not compound. I connected Claude, MCP servers, skills, Notion, Slack, notifications, and write-back into one architecture so the work moves without anyone carrying it between tools.

1

Connected operating architecture

HoneycombDeveloper Tools
Read the case study
032026–Present

Rebuilding Finance Close with AI

Finance spent a full week every month manually recalculating ARR and reconciling it against contract lines. I personally built the first version end to end, and 89% of contracts reconciled automatically end to end. The remaining 11% went to people on purpose.

89%

of contracts reconciled automatically end to end

HoneycombDeveloper Tools
Read the case study
042024–2026

Building AI Transformation from the Work Up

I started with the work: support, trust and safety, legal, marketing, internal knowledge. Each project taught me a condition AI needed to succeed. I turned those conditions into an operating model, shared infrastructure, and a community of practice, and that became the AI Ops function.

6

Functions on one roadmap

SoundCloudMusic & Media
Read the case study
052025–2026

Turning Consumer Support Into a Product

Consumer support was spread across The Weather Channel, Weather Underground, and Storm Radar, across platforms, channels, and knowledge sources, measured by throughput. I rebuilt it as one product: AI entry, human support, knowledge, routing, taxonomy, analytics, and a feedback path into Product. The AI matters here, but the system is the point.

3 brands

One support product across all of them

The Weather ChannelWeather & Data
Read the case study
062019–2024

Scaling Apptegy from 800 to 4,000 Customers

Five years taking an EdTech portfolio from startup to scale, 5X customer growth and a contribution to $20M ARR, built on operating models that outlasted my tenure.

800 → 4,000

Customers

ApptegyEdTech
Read the case study

Contact

Want the longer version?

Happy to walk through any of these in detail, including the parts that did not work the first time.

Book a call →Email me →View résumé →Other ways to reach me →