Ryan
Gillihan

I design and build practical AI tools, automations, and agents, then help teams actually adopt them.

Current role: Customer Success Operations, AI Strategy & Enablement at Smarsh. I build working software by directing AI coding tools like Claude Code, Cursor, and Codex, then get teams to actually use what I ship.

01 What I do

Built. Shipped. Adopted.

A

Applied AI

I design and build practical AI tools and agents that solve real workflow problems, directing AI coding tools to take an idea from prototype to something people use every day.

B

Automation & workflow

I build pipelines that take repetitive manual work off people's plates: consolidating inputs, auto-drafting reports, and turning fuzzy asks into repeatable processes and SOPs.

C

Enablement & adoption

I write the field guides, training, and responsible-use guidance that standardize AI-assisted work across a team, so the tools actually land instead of just launching.

02 Selected work

Things I designed and built with AI.

Enablement Tooling
SmarshPrompt engineeringResponsible AI

Internal AI enablement suite

A set of internal tools for a customer-success org: prompt-engineering and context field guides, a prompt builder, and a work-use image-generation guide. Responsible-use guidance is built in through in-product warnings and learning modules.

Standardized AI-assisted work across a 30-person CS team, with extensions built for the Enterprise and Public Sector business units.

Automation Pipeline
ReportingExec comms

Leadership reporting & GM-update automation

An automation pipeline that consolidates weekly leadership inputs and auto-drafts Business-Unit GM update emails, replacing manual report entry and assembly.

Removed manual report assembly for six department heads; monthly reporting and QBR automation queued next.

Agent
LLMData extraction

Data-processing agent

An agent that extracts data from multiple sources and uses an LLM to create structured, searchable records.

Turns scattered, unstructured inputs into clean records that can be queried.

Agent
LLMScheduled

AI news-aggregation agent

A scheduled agent that aggregates newsletters and RSS feeds into a single synthesized daily industry digest.

Replaces manual skimming with one digest, delivered on a schedule.

Infrastructure
Local LLMData security

Air-gapped local LLM machines

Multiple “off-grid” servers running local LLMs, built to power a data-security product offering where data can’t leave the box.

Gives a privacy-sensitive offering AI capability with no external data exposure.

03 About

I come from an operations and enablement background, now focused on applied AI and automation. I'm most useful where vague leadership needs have to turn into working tools that a team actually adopts.

At Smarsh I own the AI enablement roadmap for a CS organization managing a $100M+ book of business, shipping internal tooling and reporting automation while supporting executive operating rhythms. I build by directing AI coding tools like Claude Code, Cursor, and Codex, which is exactly the kind of leverage these roles are built around. B.S. in Kinesiology & Exercise Science, Oregon State University.