Nick Jaouen | Alpine Intelligence

I'm Nicholas Jaouen, the founder of Alpine Intelligence, an independent software, data, and AI engineering practice based in Jackson Hole, Wyoming.

I help organizations turn messy data, scattered documents, and complex workflows into reliable software, data pipelines, and AI-powered tools. Over the last 19 years, I've built production web applications, large-scale ETL pipelines, CRM migration systems, database-backed tools, and AI-enabled document workflows for clients in finance, healthcare, nonprofits, and other data-heavy environments.

My work sits at the intersection of software engineering, data engineering, and practical AI. I'm especially useful on projects where the problem is complex, the data is imperfect, and the solution needs both strong technical execution and good judgment.

I work across the stack using TypeScript, Python, React, Next.js, PostgreSQL, AWS, PySpark, SQL, and modern AI/RAG frameworks. I also bring a mechanical engineering background, which informs the way I think about systems, constraints, and durable solutions.

At Alpine Intelligence, I focus on building useful, maintainable technology that fits how people actually work.

Services

AI Workflow & Automation Development

I build practical AI systems that help teams search, summarize, extract, classify, and use their internal knowledge. This includes RAG systems, document ingestion pipelines, AI research workflows, structured data extraction, and human-in-the-loop review tools.

Data Engineering & ETL Pipelines

I design and build reliable data pipelines for moving, cleaning, validating, and transforming data across cloud platforms, databases, and business systems. My experience includes AWS, PySpark, SQL, PostgreSQL, Athena, and large-scale data quality workflows.

Data Migration & System Integration

I help organizations move messy, business-critical data between systems while preserving accuracy and trust. This includes CRM migrations, field mapping, deduplication, entity matching, API integrations, validation reports, and reconciliation logic.

Custom Software & Internal Tools

I build custom web applications, dashboards, admin tools, and workflow systems for teams whose needs are not well served by off-the-shelf software. I work across the stack using modern tools like TypeScript, Next.js, PostgreSQL, Spark, LangChain, and AWS.

Database & Performance Consulting

I help teams diagnose and improve slow, complex, or fragile database systems. This includes query optimization, indexing strategy, schema design, reporting performance, PostgreSQL tuning, and maintainability reviews.

Recent Projects

JHDash!

Until Recently, Jackson Hole's bus system required people to use spreadsheets on their website in order to find bus times. In the past couple years, Google Maps added the bus routes and an app called Transit started supporting Jackson Hole. I was growing frustrated with the user interfaces of both of these, so I decided to build my own, no nonsense, app for finding bus times. The idea has grown over time, and I now have a collaborator who has assisted with some important features. It started with just the bus and some weather. Now it shows the bus times, weather information, snow depth at Jackson Hole and a message of the day. Once a day, a CRON job runs that sends the day's weather, a list of events for the day, and a couple recent messages of the day to an LLM API endpoint to generate 6 message variants. The messages vary in length and tone. All are light-hearted and try to keep the user informed about what is going on in the valley on any particular day. The event data comes from "JHDash - Events!", a separate python app.

JHDash! is build using the Vercel NextJS stack, with Typescript and Node for the front and back ends, and Prisma for the Postgres database. All background photos were taken by me. Components are a mixture of custom built components and Shadcn components. It is deployed on Vercel and is available at JHDash.com

JHDash - Events!

JHDash Events! is a Python-based project utilizing LangChain and LangGraph. A LangGraph StateGraph is used to create an agentic flow that starts with a URL for a website listing events. Playwright is used as a tool by a Large Language Model (LLM) to load the page and scrape it. This information is fed into an LLM, which extracts event names, locations, dates, descriptions, and event page URLs.

The list of events is then broken into individual events, and the main pages for those events are scraped using Playwright. The data is then passed to an LLM to enrich the information already gathered about each event. Some events can be part of a series, such as a summer concert series. In that case, the LLM identifies the event as a possible series, and series candidate records are created for the individual events in that series. Those events are then researched in the same way as any other event. The final list is loaded into a Postgres database for JHDash! to access.

Throughout the process, agents are used to identify duplicate events and to standardize the names of events, organizations hosting events, and event venues. Retrieval-augmented generation (RAG) is utilized to search for existing canonical names in a Postgres database via vector search. Candidates are then sent to an LLM to decide whether they are matches. RAG is typically used as a method for augmenting the knowledge available to an LLM, but here it is used as a method for ensuring data quality.

Tavily web search is also used throughout the process to perform searches for canonical names of organizations and venues using their official websites. Tavily results are passed to an LLM for inspection.

Hard-coded rules are kept to a minimum during the process. It’s easy to fall down the rabbit hole of building complex hard-coded rules and lists of exceptions to handle every scenario and outlier. This process eschews those older methods of complex rules and functions and instead focuses on utilizing large language models to make decisions about how data should be handled. The results are surprisingly good and require far less code than a traditional deterministic approach.

The project was initially built piece by piece in several Jupyter notebooks. It was then transfered to Python scripts that can be run automatically at command line. The process runs locally on a MacMini and results are stored in the same Primsa database that JHDash! uses. There is a table for Canonical Names and aliases, along with 1536 dimensional vector embeddings that can be searched. The other table is for Events. The event table contains a composite description column which has a corresponding vector embedding that can be used for vector searches. Since vector embeddings represent the meaning of their content rather than the exact words in their content, they are ideal for finding similar names that may contain typos or be written in different ways.

View The Project on GitHub:
JHDash - Events!

Contact

Contact me: nick@alpineintelligence-llc.com