We focused on hard, boring problems; and accidentally became an AI company.

Ask ten people in your company to explain the strategy and you’ll get eight different answers.

Similarly, if you asked how the organization operates — how things are prioritized, how strategy is crafted, how opportunities are evaluated, how progress is measured — how many actually understand the operating model?

It’s hard enough to align people. And as humans, we do a pretty good job working in highly implicit operating environments.

But AI agents do not do well when the strategy and operating model are implicit. They won’t ask or clarify; they will just make poor assumptions, hallucinate, and wrongfully fill in the context gaps.

Before AI can operate in your organization, it has to understand your organization.

Your strategy, your context, your operating model.

But most of this data isn’t explicit. It’s hard enough to connect the dots across humans coordinating with other humans, but agents need an explicit map.

They need a digital twin of the organization’s way of working. A repository for strategic context; always aligning agents to the why. Providing humans and AI with better context for judgment, and shaping how they operate at scale.

That’s not why we built Dotwork. We built the platform to tackle the same boring problems we’ve dealt with for decades, but it’s becoming the platform’s biggest superpower in the age of AI. All because the same problems exist with humans, and they’re becoming even more imperative as we inevitable integrate AI into everything we do.

How we built the accidental platform for the AI era

In 2019, our company AgileCraft was acquired by Atlassian and rebranded as Jira Align. At the time, we were focused on the problem of connecting IT and ‘the business’ — supporting agile transformations at some of the largest fortune 500 companies.

When we left to start Dotwork at the end of 2022, we thought 1) we don’t know how organizations are going to operate in the future, 2) organizations operate within very different contexts, and 3) to model the complexity of ever-changing organizations, the relationships between artifacts and the ability to adapt matter more than the facade of structured hierarchies and rigid frameworks.

We built Dotwork to solve the ‘forever’ problems of connecting the dots in complex operating environments, but ended up building a platform that’s purpose-built for AI.

It’s graph-native, has an adaptive ontology, and can seamlessly query and connect data from your frontline tools without interrupting how teams operate.

With Dotwork, 90% of the value comes before using AI, but that final 10% with AI feels like magic. It solves some of the most challenging problems that plague these tools: tedious data entry, messy integrations, and context for decision support.

It also brings new problems that arise from aligning hybrid human and AI teams with strategy — many of the age-old problems still arise.

Our vision for Dotwork is to be the foundation for scaling complex operating models with strategically aligned teams of humans and agents grounded in context.

Developing building blocks for the operating models of the future means supporting the workforce of the future. A workforce that needs explicit context more than ever.

What makes Dotwork + AI different

Most tools try to force-fit strategy and operations into rigid templates, static reports, or one-size-fits-all frameworks. Dotwork takes a different approach: it was built from the ground up to model complexity, capture relationships, and adapt as organizations evolve.

Instead of assuming how your company should operate, Dotwork provides a living system that reflects how it actually does.

This flexibility is what makes it uniquely suited for the age of AI where explicit context, connected data, and adaptable structures aren’t nice-to-haves, but the foundation that allows humans and agents to align, reason, and act at scale.

Knowledge graph & ontology

Agents love graph-based systems, but legacy systems were all built on tabular data. This is a problem.

When we started Dotwork, before the launch of ChatGPT, we took our 10+ years of learnings building these systems and bet on graph. We believed relationships mattered more than records.

This means that the decision to make Dotwork the home for strategic context is an investment in the AI operating model of the future.

It means instead of strategy planning immediately becoming static in a mess of docs, decks, and spreadsheets, it’s live in Dotwork with rich connections into frontline team tools and analytics like Jira, Productboard, Looker, Snowflake, and 400+ other connections.

The flexibility for modeling complex strategic artifacts and connecting them to diverse sets of data across tools (both objects and metrics) is powered by the knowledge graph.

The Ontology gives structure to the knowledge graph — it allows you to define the types of artifacts, what they mean, the properties they have, and how they relate to each other. It allows you to shape and reshape the organization as you treat your operating model like a product; always adapting and improving.

These are new concepts in a new world that require next-generation platforms to get the most out of AI.

Explicit, structured context

As humans, we already struggle to communicate and align to a strategy; constantly asking ‘what even is the strategy?’ and ‘why does this matter?’ as we make critical decisions.

With AI, this problem is worse. In order to perform complex tasks, provide insight, and help teams make decisions, agents need even more explicit strategic context.

The problem increasingly becomes an issue of global context vs local flexibility.

Dotwork becomes a repository for strategic context. One unified source for what’s happening, what matters, who owns what, what things are aligned to, how things are funded, and what the priorities are.

When you have teams, outcomes, high-level work over time in one place, deeply connected to the tools where execution is happening, it’s a powerful command hub for agents.

Schema-flexible data integration and alignment across frontline tools

One of the hardest problems with this class of tooling is connecting the dozens of frontline tools teams rely on every day to stitch together big pictures for leaders with the least amount of manual effort.

Each system speaks its own language, enforces its own schema, and locks strategic context away in silos. Integrations typically end up as webs of brittle mappings or endless manual data entry, leaving leaders with partial, stale, or contradictory views of what’s happening.

Dotwork takes a schema-flexible, adaptive approach to integrations. We learned all the lessons the hard way building and maintaining large integrations for Jira Align.

Instead of forcing all tools into a rigid data model, Dotwork’s ontology maps to the data as it exists. Whether it’s Jira issues, Salesforce opportunities, Productboard features, custom metrics in Snowflake, or data from your BI tools, Dotwork can ingest, align, and connect that information without breaking teams’ workflows.

The knowledge graph provides the connective tissue, while the ontology provides the structure to make sense of it; all without requiring heavy re-architecture of your source systems.

The result is alignment at scale even when things are messy. Strategy is no longer buried in decks while execution lives in disconnected tools.

Dotwork creates a unified, living map where every object (a goal, an initiative, a dependency, a metric) is connected to the data that matters most. Humans get better visibility, agents get the explicit context they need to act intelligently, and organizations finally bridge the gap between strategy and execution without adding friction.

Turns out, we were building for AI all along

In October of 2022, we didn’t set out to build an AI company. We set out to solve the persistent, boring, unsolved problems of connecting strategy and execution in large enterprises. But in making the implicit explicit, capturing strategic context, aligning teams, connecting data, and wiring up operating models, we realized very quickly we were creating the foundation AI needs to actually work in the enterprise.

Dotwork isn’t just another AI wrapper or chatbot bolted onto existing workflows. It’s a platform designed for the hybrid workforce of humans and agents. A platform that scales operating context. A platform that makes AI truly enterprise-ready by giving it the strategic grounding it needs.

In hindsight, the “boring problems” were the hard problems all along. And solving them is what makes Dotwork the accidental AI company.