The first machine in this story is a PC in rural China. Hong Wang's parents bought it when he was young, though a computer was not an ordinary household object in the place or period where he grew up. They had been among the first groups to attend college after the Cultural Revolution. Education had moved the family from countryside to city, and they treated access to technology as both a privilege and a lever.
Wang learned BASIC. He played Age of Empires. He found the work of programmers far beyond his town through open-source software, one of the few clear windows into a commercial software world that was still small around him. The code came without a ticket booth. It taught him, invited him, and left a debt of the pleasant kind: one repaid by giving the next person the same chance.
Years later, that debt would acquire an octopus mascot, thousands of contributors, a foundation charter, and a company called Akuity. But the emotional logic stayed compact. Somebody had left a door open. Wang wanted to leave it open, too.
Days become minutes
Wang came to the United States for a master's degree at Carnegie Mellon University. One of his early projects involved a computation for NASA using NVIDIA graphics processors. Work that had taken days could be divided across a cluster and completed in minutes. Distributed systems gave him a way to make an intimidating problem smaller without making the ambition smaller.
Programming is so much more powerful, and so much more fun on distributed systems.Hong Wang
The line is unusually revealing because it pairs power with fun. Infrastructure writing tends to arrive wearing a hard hat. Wang talks about it with the pleasure of somebody discovering that an orchestra can play a passage no soloist could manage. Break the task apart. Coordinate the pieces. Recover the big picture.
His infrastructure career included Tintri and Springpath, but the decisive collaboration began at Applatix. There he worked with Jesse Suen and Alexander Matyushentsev on scalable production systems built with containers and Kubernetes. Wang and Suen had arrived from different geographies - Wang from China, Suen from Silicon Valley - yet their childhoods rhymed. Both had taught themselves BASIC, both played Age of Empires, and both had parents willing to drive long distances for better computer-science programs.
Open on the first commit
In 2017, the Applatix team released the project now known as Argo Workflows. The premise was to orchestrate parallel jobs on Kubernetes. The larger bet was cultural: Argo would be open from its first commit, before a private product could harden around one company's assumptions.
That decision exposed the work to inconvenient use cases, unexpected contributors, and the productive embarrassment of discovering that strangers do not share your defaults. Wang's early experience had prepared him to see this as an advantage. Public code had widened his education. Now public participation widened the product.
The clearest growth signal did not come from a dashboard. Wang remembers an early Slack channel populated mostly by maintainers fielding questions. Then members began answering one another. The founders were no longer the switchboard for every useful exchange. The room had learned to run itself.
At first, the team could not even see much of the room. Open-source software does not send a customer list when somebody downloads it. Word drifted back that Alibaba, BlackRock, and Google were using and contributing to Argo. At KubeCon in 2017, Kubernetes co-creator Joe Beda stopped at the booth and showed interest in their Kubernetes-native workflow engine. The encounter gave the team an external calibration point: the problem that felt urgent from inside their startup also made sense to people who had shaped the platform beneath it.
When the community is willing to showcase what they're doing, that makes me feel like I'm building something amazing.Hong Wang
There was serendipity, too. Applatix was acquired by Intuit in 2018. Instead of being folded away as acquisition archaeology, Argo gained a demanding production environment. Wang and Suen led work across Kubernetes platform and developer-experience teams. They did not merely hand out a tool; they operated it with internal service commitments, guidance, and support.
Operating at that scale is a useful antidote to founder poetry. Every elegant abstraction eventually meets an upgrade, an outage, a confused user, and a Friday afternoon. Argo CD emerged from the need for a declarative continuous-delivery system built specifically for Kubernetes. Git held the intended state. Argo compared it with the live system, exposed drift, and reconciled the difference.
The ship leaves the company
Argo took its name from the ship that carried Jason and the Argonauts in Greek mythology. An argonaut is also a pelagic octopus, which explains the mascot and supplies a convenient metaphor: many arms coordinating many tasks. The mythology is charming. The governance mattered more.
The project entered the Cloud Native Computing Foundation as an incubating project in 2020. That neutral home made it possible for vendors and users to contribute without treating Argo as the private garden of one employer. In 2022, it graduated, a marker of maturity, adoption, and sustainable stewardship.
Graduation also revealed how far the job had traveled from engineering. Wang, Suen, and Matyushentsev have written about learning to wear unfamiliar hats: documentation, conference talks, blog posts, governance, and the patient diplomacy of a project used by competing vendors. Code can begin a community, but it cannot host a conference, explain a breaking change, or persuade an anxious operator that the upgrade path is sound. A maintainer eventually discovers that clarity is infrastructure too.
By then, Wang and Suen had reached an awkwardly fortunate conclusion. Argo had too much momentum to remain a spare-time responsibility. Large organizations wanted support, predictable operations, and tooling around the open-source core. In 2021 they formed Akuity. The company announced a $4.5 million seed round, then added a $20 million Series A in May 2022.
Solve the internal problem. Open the first commit. Let outside use cases reshape the work. Put neutral governance around the community. Build the company only when trust creates a real support problem.
This sequence is not merely tidy history. It is the strategic core of Wang's story. Akuity did not invent a community because a distribution slide demanded one. The business followed a project whose utility had already escaped its creators. Commercialization carried an obligation: improve the paid product without enclosing the commons that made it possible.
What happens after deployment?
The next problem had been waiting since the early Argo CD days. A deployment tool can reconcile one environment beautifully. Real software travels through several: development, testing, staging, and one or many production destinations. It needs approvals, verification, rollbacks, and a visible answer to the question, "Where is this change now?"
Resources were scarce, so the team focused Argo CD on deployment and deferred the journey between environments. At Akuity, the old idea became Kargo, an open-source continuous-promotion project. If Argo CD is the dependable harbor master for an environment, Kargo is the route plan between ports.
In March 2026, Wang reported that Akuity had more than 100 customers and had processed over 43 million successful deployments in 2025, nearly ten times the prior year's volume. The company tied much of that acceleration to artificial intelligence: more code, more fixes, more branches, and consequently more things asking to enter production.
Wang's response to the AI coding boom is pleasantly infrastructural. If creation becomes cheaper, coordination becomes more valuable. Engineers may spend less time typing structures and APIs and more time on architecture, context, security, and judgment. The systems downstream must absorb the new volume without turning speed into roulette.
In the AI era, GitOps becomes more foundational. Everything will be code, everything will be declarative, and everything will be managed that way.Hong Wang
Akuity has added AI-assisted diagnosis and remediation, while Wang continues to insist on the importance of context and guardrails. The aspiration is not to make operations mysterious. It is to let humans define intent clearly enough that both software and machines can act safely.
The door stays open
Wang's career is easy to summarize as a procession of scale: from one PC to distributed GPUs, from a tiny Slack room to a global project, from a handful of clusters to tens of millions of deployments. Yet the more interesting continuity is access. His parents understood that access to education and technology could alter a family's coordinates. Open source gave him access to a wider engineering culture. Argo offered that same widening to others.
His ambition for the work has stayed broad but legible. He has described Kubernetes as a common API for cloud computing and imagined Akuity and Argo as a lingua franca for the automation around it. The phrase is grander than the daily labor, which is precisely why the daily labor matters. Universal languages are assembled out of ordinary verbs, good documentation, boring reliability, and enough humility to accept an accent you did not design.
There is no clean separation here between gratitude and ambition. The wish to pay back an early gift produced work that also became a career, a community, and a venture-backed company. Idealism learned to carry a pager. Capital learned to share a repository. The octopus acquired enterprise service agreements.
That is the durable lesson in Hong Wang's route. Communities compound when the creator stops needing to be the center of every exchange. Products mature when they meet other people's inconvenient realities. Companies earn a place when they support value that already exists. The ship matters, but so does the decision to let more people aboard.