Case Studies

Real results from real work.

How we've helped firms shape AI initiatives and put AI and automation to work.

Internal Operations

AI-Powered Business Operating System

01
The Challenge
My Notion workspace was already doing a lot of heavy lifting. Light CRM, task management, meeting notes, project tracking. All relational databases that link together nicely. Adding a task and connecting it to a project and client is fairly simple. Despite the structure being solid, I was still the one doing the work. Adding tasks. Updating the CRM weekly. Pulling action items out of meeting notes. Adjusting project timelines. All those little things that individually take 5-15 minutes but add up to hours across a week.
02
The Solution
Built an integrated system using Claude Code and Antigravity (Google's AI IDE) that connects directly to Notion, Google Drive, and Fireflies. Fireflies records meetings and auto-pushes transcripts to Notion. From there, an AI layer analyzes each transcript: action items get extracted and added to the right project in the tasks database. New contacts get created in the CRM. Existing contacts get updated with last interaction dates. Documents route to the correct Google Drive folders. The AI brain knows where things go because I built it with clear guardrails. The whole workspace becomes queryable. It's easy to quickly ask the AI a question and get an instant answer.
03
The Impact
Roughly 15+ hours saved per week. Very little post-meeting admin left. I still double-check things and fill in a few details, but the system handles the bulk of the busywork. It's an ongoing build that keeps getting smarter as I add more workflows.
Accounting Firms

Firm Operations Assessment

Completed for multiple ~$10M revenue accounting firms.

01
The Challenge
Firms know they need to move on AI and automation, but most don't have a clear picture of where to start. Jumping straight into implementations without understanding current processes is risky. You end up investing in the wrong areas or missing the opportunities that would actually move the needle.
02
The Solution
Met with firm leaders to understand the current tech stack, service lines, and day-to-day operations. Conducted interviews with staff across the organization to map out processes and identify opportunity areas. Delivered a comprehensive report covering process optimizations, AI opportunities, automations, agentic workflows, and software recommendations. Presented findings to leadership with a clear view of the landscape and a stack-ranked list of priorities.
03
The Impact
These assessments give firms a strong foundation before pushing forward with process improvements and AI. Clear picture of current operations, with high-impact opportunities identified and stack-ranked for the firm to pursue. Several of these projects have led to automation builds and AI implementations.
Marketing Agency

Brand-Specific Content Agents

01
The Challenge
Content production was eating up the team's time. Hours spent brainstorming topics, drafting posts, reformatting for different platforms, and adapting content to match each client's voice. Even with AI tools like ChatGPT, there was constant back-and-forth to get the tone, style, and brand guidelines right.
02
The Solution
Built a system where each client's brand guidelines, writing style, tone, and SEO/GEO/AEO constraints are managed in one place by any team member. AI agents connect to those guidelines to produce blog posts based on identified content topics. Posts are prepared and pushed to a review and scheduling platform that fits into their existing workflow.
03
The Impact
Estimated 10-20 hours saved per week compared to fully manual content production. Even compared to using standard AI chatbots with lots of prompting back-and-forth, the system saves at least 5-10 hours weekly. The workflow is now more scalable as the agency takes on new clients.
Advisory Firm

Automated S Corp Compensation Reports

01
The Challenge
The firm produces S Corp reasonable compensation reports for clients. Each report requires research, rate analysis, and formatted documentation. Done manually, a single report takes 5+ hours of work. That's a significant time investment that limits how many reports the firm can handle and eats into margins.
02
The Solution
Built a system that runs the backend for the firm's report production. Form responses log to an Airtable database, then flow through proprietary AI workflows that research the inputs and determine specific rates to apply. A Google Docs template generates a formatted draft report ready for the team to review. The entire process runs automatically.
03
The Impact
Reports are prepared within minutes of form submission. What would take 5+ hours manually now happens automatically. The system significantly improves the firm's margins and makes the service far more scalable.
Internal Operations

Newsletter Production System

01
The Challenge
Producing a weekly newsletter requires hours of research, content curation, and writing. Scanning YouTube, podcasts, Reddit, LinkedIn, Twitter/X, and newsletters for relevant stories. Remembering news items heard throughout the week. Then writing it all up in a consistent format. The work adds up fast.
02
The Solution
Built a multi-system pipeline that handles research and drafting while keeping a human-in-the-loop for final decisions. Content scrapers pull from YouTube, podcasts, Reddit, LinkedIn, Twitter/X, and Gmail newsletters weekly. A Slack agent lets me log news items I hear throughout the week. An AI layer analyzes the content and surfaces the 8-12 most relevant stories for the audience. After I select the 3-4 best, the system drafts each story and populates a formatted Google Doc ready for final editing.
03
The Impact
20+ hours saved per week on manual research, writing, and content curation. The system surfaces timely, relevant stories I might have missed. But I still make all final decisions and write the actual newsletter. Human judgment, AI-powered workflow.