We are Atlas Axis, a startup venture incubated by Sky (formerly MakerDAO). As one of the earliest DAOs on Ethereum, Sky was launched in 2014 and became the first DeFi application to earn significant adoption. Today, the Sky Ecosystem stewards the longest-running protocol on Ethereum with a TVL of over $5 billion.
To achieve a resilient, decentralized and unbiased world currency, Sky is undergoing a radical structural, organizational and cultural transformation which it has dubbed “Endgame.” The heart of Endgame is the Atlas, the AI-enhanced governance ruleset powering the entire Sky Ecosystem. Atlas Axis has been entrusted with developing the Atlas. Our day to day work has made apparent the stark limitations of existing tools in an AI-native world.
LLMs offer tantalizing glimpses of an unprecedented cognitive partnership, yet the software paradigms we inhabit often constrain this potential, especially for deep, nuanced knowledge work. The dream of truly personal, malleable software, environments that adapt to individual intellects and workflows, rather than forcing conformity to pre-defined applications, remains largely unfulfilled.
How do we architect and build computational environments that allow individuals to fluidly shape their tools? How do we integrate AI as a true collaborator, augmenting versus outsourcing our intelligence? How do we move beyond rigid applications to create malleable substrates where knowledge can be explored, connected, and built upon in ways that are as dynamic and personal as thought itself?
The challenge is immense, yet the opportunity for impact is even greater. Consider the rise of Cursor, Microsoft already had the models and the data, but Copilot was shipped as technical debt, a cautious plug-in layered onto an existing editor. In this new landscape, the most defensible moat isn’t necessarily vast datasets or existing market share, but rather a profound understanding of the creative problem faced by the expert-user and an innovative, opinionated take on what work should look and feel like, a take that harnesses on first principles the full potential of AI.
Cursor treated “AI coding” first and foremost as a product-design problem, asking “How can an IDE make an expert-user tangibly faster and more effective in their work, with as little cognitive friction as possible?”. We are asking the same question for ourselves, and more generally, how it can empower knowledge workers engaged in deep analytic work.
We are seeking a visionary Lead Engineer, to be a foundational architect and builder in this quest. You won’t just be implementing features, you’ll be tackling fundamental challenges in how software mediates thought, how AI augments human capability, and how complex information can be made intuitively navigable and creatively extensible.
Your initial focus will be building automated workflows and data systems (e.g., synchronizing complex, hierarchical content between platforms like Notion and Supabase) that serve as both critical infrastructure and early explorations into these deeper challenges of structure, change, and representation.
This role is for someone who is not just an excellent engineer, but a systems thinker, passionate about the future of knowledge work, eager to explore the uncharted territory at the intersection of AI, HCI, and personal computing. If you are driven to build systems that empower individuals to think better and create more profoundly, we want to hear from you!
"We are also open to onboarding on a short-term/per-project basis an engineer who can lead Phase 1; while continuing to search for a suitable long-term team member who can lead Phase 2."
-
Phase 1 — Atlas Workflow Automation (0-6 months): Building the Engine
-
Lead the design and implementation of the “Atlas Workflow Automation v2” system, focusing on robust data synchronization between Notion and our backend, precise change-tracking for hierarchical content, and automated governance proposal generation.
-
Develop and own the reliability of these data pipelines, including implementing automated testing, comprehensive monitoring, and robust rollback/repair tooling.
-
Architect and build core backend services and data models with an emphasis on event-driven architecture.
Success in this phase includes: planned Automations running end-to-end; data loss < 0.1%; proposal generation in < 5 min; recover-from-failure < 15 min.
-
-
Phase 2 — Knowledge-Work IDE Foundations (6-12 months): Crafting the Future
-
Generalize the primitives and patterns from Phase 1 to create an extensible platform for a future AI-augmented, malleable Knowledge-Work IDE.
-
Research, prototype, and iterate on new cognitive interfaces (e.g., graph-outline hybrids, spatial canvases, dynamic lenses) and plugin architectures, collaborating with design and AI/research partners.
-
Explore and integrate AI/LLM capabilities at a first-principles, systemic level to reshape knowledge interaction and tool creation.
-
5+ yrs building production back-end / server-side systems, with strong proficiency in JavaScript/TypeScript (Node.js) AND demonstrated experience with at least one of Python, Go, or Rust/Java.
-
Expert-level SQL proficiency and deep experience designing and working with relational databases to manage complex, hierarchical data structures. You must have a proven ability to model parent-child relationships, ensure data integrity, and write sophisticated queries for data retrieval, transformation, and change tracking. Significant hands-on experience with PostgreSQL is a must.
-
Extensive experience with complex third-party REST APIs, including demonstrable, in-depth experience with at least one major content or productivity platform API (e.g., Notion, Google Workspace, Airtable). Proficiency with, or ability to quickly ramp up on, Notion API production.
-
Proficiency with, or ability to quick ramp up on, design and implementation of diffing algorithms or robust change-tracking logic for structured/hierarchical data.
-
Solid grasp of API design principles (REST/GraphQL), secure authentication patterns, and secret management.
-
Experience with event-driven architecture, containerization (Docker), modern CI/CD practices, and pipeline observability (automated tests, monitoring).
-
Deep into AI-assisted coding and eager to test/taste all the new models and tools
-
Experience with Model Context Protocol (MCP) servers.
-
Experience with PostgreSQL-based Backend-as-a-Service (BaaS) platforms (like Supabase) or building custom backend services on PostgreSQL that expose data via APIs is a strong plus, particularly if you’ve leveraged platform-specific features (e.g., advanced functions, RLS) for automation, security, or data synchronization in such an environment.
-
Vision, Mindset & Collaborative Spirit
-
Deep interest in AI/ML, particularly LLMs - obsessed with their potential to transform software, computing and knowledge work.
-
Love to nerd out on “tools for thought,” software malleability, and the history and future of personal computing.
-
Experience or strong interest in designing extensible platforms, frameworks, or systems with rich plugin architectures or APIs for customization.
-
Excellent technical communication and a collaborative, first-principles thinking approach.
-
Resourceful, self-directed, and comfortable navigating the ambiguity inherent in exploring new frontiers.
