AI Tooling & Custom Builds
- Category
- Data Factory
- Description
- 1-on-1 AI advisory that scales into full builds. Start by folding LLM agents and AI workflows into your day-to-day, then extend into custom software.
- Estimated duration
- 1-2 hours per session
- Order
- 2
- Parent
- Empty
- Pricing
- $75/hr
Are you trying to use AI agents in your day-to-day and don't know where to start?
Do you work in healthcare, or any field with a lot of data to wrangle? Are you trying to turn that data into published insights?
No more...
- Spending hours formatting the same recurring deliverables for stakeholder reviews
- Copy-pasting data into spreadsheets to track trends
- Manually compiling competitive landscape reports across your product portfolio
- Watching colleagues talk about "leveraging AI" in meetings without anyone showing concrete results
- Wondering if another certification is really what's going to set you apart
What a build can look like
Not every project is a regulatory platform. Most start small, and small is often all you need:
- A script that cleans, merges, or analyzes a messy spreadsheet or export, turning an hour of manual work into one command.
- A dashboard or visualization that makes a pile of data legible, filterable, and shareable.
- A simple website or landing page. Fast, clean, and live in days, not months.
- A workflow that connects two tools: a form that auto-populates a sheet, a sheet that triggers an email, a folder that syncs itself.
- Something that talks to an API: pulling, pushing, or syncing data between services on a schedule, so you stop doing it by hand.
- A chatbot or internal tool powered by Claude (or another LLM): answering questions over your own documents, drafting in your own voice.
Bigger builds, like the healthcare projects below, grow from exactly these.
How it works
- You share your role and what's eating your time. Product reports? Regulatory tracking? Portfolio analysis? We start with what's real for you.
- We pick 1-2 workflows that are high-effort and repeatable. These are the best candidates for AI, tasks you do weekly or monthly that follow a pattern.
- You document the process as if training someone new. This is the key insight, the same clarity that makes a good SOP makes a great AI prompt.
- We feed it to an LLM agent and iterate live. You'll see your workflow get automated in real time, and learn the prompting patterns to replicate it.
- You leave with a working artifact and a repeatable playbook. A prototype, a dashboard, an analysis, something you can share, publish, or present to leadership.
Pricing
Book a session
FAQ
Do I need to know how to code?
No. The whole point is to meet you where you are. If you can write a clear email, you can write a good AI prompt. We focus on documenting your process clearly. That's the skill that transfers to working with LLM agents, not coding.
What tools or software do I need?
Just a browser. We'll use tools like Claude, ChatGPT, or other LLM agents depending on your use case. I'll recommend the right tool for your specific workflow. You don't need to buy anything in advance.
Is this specific to healthcare, or can it apply to other industries?
My niche is healthcare, biotech, and life sciences. That's where I work every day and where I can offer the most specific guidance. The process of identifying workflows and automating them with AI applies anywhere, but you'll get the most value if your work touches healthcare data, regulatory, or product analysis.
What if I'm not sure which workflow to start with?
That's exactly what the first 15 minutes are for. I'll ask you about your week: what you spend the most time on, what feels repetitive, what you wish you could hand off. We'll find the right starting point together.
How is this different from an AI course or certification?
A course teaches you concepts. This session builds you something real. You won't leave with a certificate. You'll leave with a working tool and the knowledge to build the next one yourself. Certifications tell employers you're interested in a space. Artifacts tell them you can deliver value in it.
Personal app build-outs
Apps I have designed and shipped end to end, proof of the custom builds this service delivers.
FDA CRL Analyzer
Analyzes 202 FDA Complete Response Letters (2020-2024) to extract regulatory intelligence, deficiency patterns, therapeutic area trends, and sentiment. Uses DuckDB for fast SQL analytics and Parquet for efficient storage. 8 deficiency categories, 7 therapeutic areas.
SaMD Clearance Visualization
Interactive visualization of FDA 510(k) clearances for Software as a Medical Device (SaMD). Explores regulatory trends, predicate device networks, and clearance timelines across product codes.
Notion-Based Glucose Dashboard
A visualization of 14 days of my glucose data (or yours if you want). Now includes Part 2: an Abbott Lingo vs Dexcom Stelo comparison, built with a Python pipeline and explored via a Notion MCP server.
CADe/CADt Regulatory Sentiment Analysis
An interactive intelligence brief over 20 FDA docket comments: sentiment distribution, stakeholder breakdowns, filterable submitter cards, recurring themes, and executive takeaways.
DDx Daily
A daily clinical-reasoning game for USMLE Step 1 and Step 2 prep. Instead of multiple-choice recall, you interview an AI standardized patient grounded on a hidden case fact sheet, order labs and imaging, then name the diagnosis, scored on accuracy AND efficiency (fewer, sharper questions earn more points). Built with Astro 7, React 19, and Turso, a Gemini-powered patient, a MedQA case-import pipeline, and real PubMed-grounded references.
LinkedIn Notes
A Chrome extension that adds private notes and color-coded tags to LinkedIn profiles. All data stays local in the browser, no servers, no accounts, no tracking. Includes a CRM dashboard, quick search popup, and JSON export. Built with WXT + React.
Partners
(Data Camp)Out of Pocket is a healthcare education and media company. Its Data Camp brings operators, clinicians, and builders together to learn healthcare by doing.
Service offerings
- Healthcare business courses and bootcamps
- Data Camp live learning events
- Industry newsletter and media
My role
Built Pyaar Trivia, a custom AI-assisted trivia platform, and ran the Out of Pocket Data Camp game as a live implementation of it. Play the game at trivia.pyaarproject.org/out-of-pocket.
(Open source)Kitware builds open-source scientific computing and medical imaging platforms, including VolView, ITK, and VTK, used across research and clinical software.
Service offerings
- Open-source medical imaging (VolView, ITK, VTK)
- Scientific data and 3D visualization
- Custom software and R&D services
- Computer vision and imaging AI
My role
Built pyaar-dicom, a local-first DICOM annotation tool, on top of Kitware's open-source VolView viewer.
(Music)Spoonerism is the music and creative team behind spoonerism.fm, running releases, community, and day-to-day operations.
Service offerings
- Music releases and radio
- Community and events
- Marketing and brand
My role
Built a customized fork of Pando for the Spoonerism team to run their marketing and operations management. See the team at spoonerism.fm.