For three decades, the quiet Berkeley firm that has been teaching data how to draw itself - one graph, one layout, one untangled mess at a time.
The mark of a 1992 software company that never needed an outside dollar to grow. Berkeley, California.
Somewhere right now, an analyst is staring at a spreadsheet of ten thousand rows and quietly losing the will to live. A fraud ring is hiding in there. So is a broken supply chain and a telecom outage waiting to happen. The rows know everything and reveal nothing. This is the exact problem Tom Sawyer Software has been solving since the first Bush administration: turning relationships you cannot see into pictures you cannot unsee.
Most data tools are obsessed with the things. Tom Sawyer Software has always been more interested in the lines between them - the edges, the connections, the suspicious little arrows that point from one bank account to another. When the company opened in 1992, "graph visualization" was not a market. It was barely a phrase. The founders bet that businesses were drowning in data and would eventually need a way to read it as a map instead of a ledger. They were early by about twenty years.
The name, charmingly, has nothing to do with whitewashed fences. It comes from the Rush song "Tom Sawyer" - a detail that tells you the founders were engineers with a record collection, not marketers chasing a metaphor. It stuck. Three decades later it is one of the more recognizable names in a field full of acronyms.
"Bringing clarity to real-world business challenges through effective graph visualization and analysis."- The company mission, unchanged in spirit since 1992
Brendan Madden trained as a physicist - a Bachelor of Science in Applied and Engineering Physics from Cornell, a physics degree from Ithaca College, graduate work in electrical engineering at Arizona State. Before founding the company he did time at IBM's Thomas J. Watson Research Center and the Santa Teresa Laboratory. Then in 1992 he picked one problem - how to draw a graph well - and has spent roughly forty years on it without getting bored. That is either remarkable focus or a very specific kind of stubbornness. In graph theory, the two are hard to tell apart.
His co-founder, Patrick, remains with the company as Principal Solutions Architect. The result of that durability is a company with institutional memory measured in decades, not funding rounds. While competitors raised, pivoted, and got acquired, Tom Sawyer Software kept shipping layout algorithms.
A low-code platform for building custom graph and data visualization apps - with drawing, layout, map, timeline, and chart views. The thing analysts build on.
Launched in 2024 for people who want to explore connected data interactively without building anything first. Click, follow, discover.
A browser-based viewer that auto-visualizes SysML v2 system models. Sold on AWS Marketplace; release 2.1 tracks the 2026-03 reference spec.
Point it at a graph database and actually see what is in there - nodes, edges, patterns - instead of writing one more query.
The original component software - the Graph Layout Toolkit and Graph Editor Toolkit - that other software vendors embed in their own products.
The customer list reads like a cross-section of the modern economy: telecommunications, financial services, defense and intelligence, manufacturing, and government. Startups use it. Fortune 1000 companies use it. The same engine that maps a fraud network can lay out a supply chain, a digital twin, a cyber-threat map, or a systems model. Hundreds of use cases, one underlying obsession - clarity.
// illustrative mix of customer industries, not audited figures
The first pure graph technology company - applying graph expertise to hundreds of use cases and technology stacks, from startups to the Fortune 1000.- On three decades of staying in one lane, on purpose
Back to that analyst and their ten thousand rows. With a Tom Sawyer graph in front of them, the spreadsheet stops being a wall of numbers and becomes a map. The fraud ring is a cluster of fat lines. The broken supply chain is a missing edge. The thing they were supposed to find has been quietly waiting to be drawn - and now it has been. No new data, no new database. Just lines, between things, finally visible. Which was the whole point in 1992, and still is.