这一切都始于一些愤怒的电子邮件。早在 2009,，一群研究人员就开始发泄他们对编程语言的不满，这些编程语言旨在帮助科学家和其他研究人员在不学习如何编码的情况下执行复杂的数学运算和统计模拟。这些编程语言既僵化又缓慢。如果科学家构建了真正有效的东西,，他们’d需要用另一种语言重写整个程序，只是为了更快地运行它。
It all started with some exasperated emails. Back in 2009, a group of researchers began venting their frustration with the programming languages designed to help scientists and other researchers perform complex mathematical operations and statistical simulations without learning how to code. These programming languages were rigid and slow. If scientists built something that really worked, they’d need to rewrite the entire program in another language just to run it more quickly.
这些电子邮件变成了麻省理工学院的一个研究项目，其使命是构建一种名为 Julia, 的易于使用的, 高性能编程语言，该语言专为科学研究, 数据分析, 和建模复杂系统（例如喷气发动机, 药品, 金融市场, 和机器人, 等）而设计。
The emails turned into a research project at MIT with the mission of building an easy-to-use, high-performance programming language called Julia, which is designed for scientific research, data analysis, and modeling complex systems such as jet engines, drugs, financial markets, and robots, to name a few examples.
该研究项目变成了 MIT, 的一个实验室，而该实验室又变成了 JuliaHub 公司。一路走来, Julia 在科学家, 工程师, 数学家, 等人中赢得了忠实的追随者。如今, 这种免费开源语言拥有超过 100 万用户,，其中包括在世界各地数千家公司和大学工作的人员。
That research project turned into a lab at MIT, and the lab turned into the company JuliaHub. Along the way, Julia gained a loyal following among scientists, engineers, mathematicians, and others. Today, the free and open-source language counts more than 1 million users, including people working in thousands of companies and universities around the world.
稍微夸张地说，Julia 已被用来模拟太阳下, 的一切，从微小原子的行为到半导体, 神经网络, 赛车, 和飞机。天文学家还使用 Julia 来对黑洞进行成像，以研究太阳,以外的许多领域。
It is only a slight exaggeration to say Julia has been used to model everything under the sun, from the behavior of tiny atoms to semiconductors, neural networks, race cars, and airplanes. It has also been used to study much beyond the sun, with astronomers using Julia for imaging black holes.
Julia的 的秘密在于它根据所使用的数据类型编译代码的方式。这样的“即时编译”使得Julia比其他数值编程语言更快、更灵活。
Julia的 secret sauce is in the way it compiles code depending on the type of data being used. Such “just-in-time compilation” makes Julia faster and more flexible than other numerical programming languages.
“科学家和工程师不是程序员。与科学家,工程师,和程序员组成的多学科团队一起构建科学应用程序具有挑战性,” JuliaHub 联合创始人兼首席执行官 Viral Shah 表示。 “我们问:如果你能为科学家和工程师配备一种编程语言，让他们能够高水平地表达他们的想法，同时获得出色的软件性能?”会怎样？
“Scientists and engineers are not programmers. Building scientific applications with multidisciplinary teams of scientists, engineers, and programmers is challenging,” JuliaHub co-founder and CEO Viral Shah says. “We asked: What if you could equip the scientists and engineers with a programming language that allowed them to express their ideas at a high level and also get great software performance?”
让非程序员轻松编程一直是 JuliaHub的 创始人, 的北极星，其中包括 Julia 联合创始人 Shah, 麻省理工学院数学教授 Alan Edelman, Jeff Bezanson SM ’12, 博士 ’15, 和前麻省理工学院研究科学家 Stefan Karpinski。
Making programming easy for non-programmers has been a north star for JuliaHub的 founders, who include Julia co-creators Shah, MIT professor of mathematics Alan Edelman, Jeff Bezanson SM ’12, PhD ’15, and former MIT research scientist Stefan Karpinski.
今年 4 月，, JuliaHub的 团队在这个方向上又迈出了一大步，推出了最新版本的 Dyad 3.0, AI 平台，帮助工程团队加速复杂物理系统的开发，如火箭, 热泵, 和卫星。工程师已经在使用 Dyad 来指导自主 AI 代理进行物理模拟, 安全分析, 质量控制, 等等。
In April, JuliaHub的 team took another big step in that direction with the launch of Dyad 3.0, the latest version of its AI platform to help engineering teams accelerate the development of complex physical systems like rockets, heat pumps, and satellites. Engineers are already using Dyad to direct autonomous AI agents as they work through physics simulations, safety analyses, quality controls, and more.
“使用 Dyad 3.0,，您可以上传数据和设计文档，系统将为您设计整架飞机,” Shah 说。 “与波音,等客户合作，我们正在为工程师构建代理硬件设计能力。简单地说,你想说,‘好吧，计算机,为我建造一架飞机’;上传设计文档;并拥有所有物理的系统帐户,编译所有代码,验证所有内容,并以代理方式构建整个设计。”
“With Dyad 3.0, you can upload data and design documents and the system will design an entire aircraft for you,” Shah says. “Working with customers like Boeing, we are building agentic hardware design capabilities for engineers. Simplistically, you want to say, ‘Okay computer, build me a plane’; upload the design documents; and have the system account for all the physics, compile all the code, verify everything, and build the entire design agentically.”
核心理念是创建一个高性能平台，在工程,、科学, 和数学应用方面表现出色。 Shah 表示，在 Julia, 之前，科学家和工程师要么必须雇人为他们构建软件，要么接受为他们设计的少数编程语言的缓慢性能。
The core idea was to create a high-performance platform that would excel at engineering, scientific, and mathematics applications. Shah says before Julia, scientists and engineers would either have to hire someone to build software for them or accept the slow performance of the few programming languages designed for them.
“我们希望创建像 Python 或 MATLAB 一样易于使用但又像 C 编程语言一样快的东西,” Shah 说。 “我们为自己构建了 Julia。”
“We wanted to create something as easy to use as Python or MATLAB but as fast as the C programming language,” Shah says. “We built Julia for ourselves.”
埃德尔曼说，起初,研究人员并没有’认为有人会想要他们的创作。
Edelman says at first, the researchers didn’t think anyone would want their creation.
“我们认为需要 10 年时间才会有人感兴趣,，但我们说, ‘耐心是一种美德,，所以让 去做吧,’” Edelman 回忆道。
“We figured it would take 10 years before anyone was interested, but we said, ‘Patience is a virtue, so let的 do it,’” Edelman recalls.
麻省理工学院的研究人员于 2012 年在一篇博客文章中宣布了 Julia。他们很快意识到许多其他研究人员也有同样的挫败感。
The MIT researchers announced Julia with a blog post in 2012. They quickly realized many other researchers shared their frustration.
“当我们第一次开始,时，我们的目标是交互式研究工作流程,，但越来越多的人将它用于一切,” Bezanson 说。 “现在，随着我们与用户一起发展，我们’正在将整个语言堆栈转移到更小的,嵌入式设备上。”
“When we first started, we were targeting interactive research workflows, but increasingly people are using it for everything,” Bezanson says. “Now we’re moving the whole stack of the language onto smaller, embedded devices as we evolve with our users.”
从早期开始，, Edelman 就为来自麻省理工学院几乎每个系的学生开设了 Julia 课程。今天,，他经常了解到学生在报名上课时已经在使用 Julia，其应用范围广泛，如机器人, 天文学, 物理模拟, 和金融。
Since those early days, Edelman has taught a class on Julia with students from nearly every department at MIT. Today, he often learns students are already using Julia when they enroll in the class for applications as wide ranging as robotics, astronomy, physics simulations, and finance.
“研究人员找到我并说, ‘我告诉我的主管我’m使用Julia，因为它’快,，但不’t告诉他们我’m使用Julia，因为它’真的 fun,’” 埃德尔曼说。 “关键是 Julia的 抽象。很多时候，一种编码语言会迫使您解决您’正在思考的一个问题。 Julia的 语言使您 不仅能解决您 正在思考, 的问题，还能解决世界各地其他 人的问题。它鼓励您更广泛地解决问题。”
“Researchers come up to me and say, ‘I tell my supervisor I’m using Julia because it的 fast, but don’t tell them I’m using Julia because it的 really fun,’” Edelman says. “The key thing is Julia的 abstractions. A lot of times a coding language forces you to solve the one problem you’re thinking about. Julia的 language makes it so you’re solving not only the problem you’re thinking about, but other people的 problems around the world too. It encourages you to solve problems more generally.”
随着 Julia 越来越受欢迎，世界各地的, 研究人员开始向 Julia 团队寻求支持。到 2015,，需求变得足够强劲，他们决定启动 JuliaHub 并通过公司全职帮助用户。他们得到了麻省理工学院德什潘德技术创新中心和麻省理工学院其他机构的支持，使公司得以起步。
As Julia gained popularity, researchers around the world started asking the Julia team for support. By 2015, the demand became strong enough that they decided to start JuliaHub and help users through the company full-time. They received support from the MIT Deshpande Center for Technological Innovation and others at MIT to get the company off the ground.
JuliaHub的工作已经从简单地帮助用户发展到更广泛地推进语言的发展。那的 为 Julia的 忠实用户提供了一系列令人印象深刻的创作。 Julia 已用于模拟计算机电路, 检测健康差异, 模型全球气候和海洋, 分析大脑活动, 等等。
JuliaHub的 work has evolved from simply helping users to advancing the language more generally. That的 powered an impressive list of creations from Julia的 loyal users. Julia has been used to simulate computer circuits, detect health disparities, model global climates and oceans, analyze brain activity, and more.
有人在 Julia, 中构建了药物建模平台后，它被用来加速 Moderna的 Covid-19 疫苗的开发。在另一个案例中，, 研究人员使用 Julia 创建了一个避免飞机相撞的程序。他们发现它的运行速度比基于 Python 构建的早期版本快约 50 倍。 Meta 的工程师使用 Julia 为 WhatsApp的 40 亿用户开发了更好的音频编解码器。
After someone built a pharmaceutical modeling platform in Julia, it was used to accelerate development of Moderna的 Covid-19 vaccine. In another case, researchers used Julia to create a program for avoiding aircraft collisions. They found it ran about 50 times faster than an earlier version built on Python. Engineers at Meta used Julia to develop a better audio codec for WhatsApp的 4 billion users.
“多年来，我们’已经看到工业,政府,和学术用户用Julia语言做各种有趣的事情,” Edelman说。 “It的 在很多方面都让我们感到惊讶, 人们使用它的范围很广。”
“Over the years we’ve seen industrial, government, and academic users doing all kinds of interesting things with the Julia language,” Edelman says. “It的 honestly surprised us in many ways, the wide-ranging things people are using it for.”
JuliaHub 于 2025 年 6 月推出了 Dyad 1.0，作为加速编程的研究代理，并于 12 月推出了 Dyad 2.0。创始人相信 Dyad 3.0 代表了设计复杂系统的能力和自主权的新水平。
JuliaHub launched Dyad 1.0 in June of 2025 as a research agent to accelerate programming and Dyad 2.0 in December. The founders believe Dyad 3.0 represents a new level of ability and autonomy for designing complex systems.
“ Dyad 的一个重要特点是它是一个物理编译器，因此强制执行物理定律,” Shah 解释道。 “一般人工智能系统经常以违反物理定律的方式解决物理问题。当使用 Dyad 代理, 时，它将检测此类违规行为并引导代理朝着物理上正确的解决方案方向发展。我们预计它将把产品工程的设计时间缩短几个数量级,，从而使数月的工作在数小时内完成。”
“One important thing about Dyad is that it is a physics compiler and hence enforces physical laws,” Shah explains. “General AI systems often solve physical problems in ways that violate physical laws. When using the Dyad agent, it will detect such violations and guide the agent in the direction of the physically correct solution. We expect it will decrease design times in product engineering by orders of magnitude, leading to months of work being accomplished in hours.”
爱德曼通过他的课堂了解朱莉娅的影响的一种方式。一名学生最近使用 Dyad 来模拟机器人如何在太空中移动。另一个用它来制造火箭发动机。
One way Edelman sees the impact of Julia is through his class. One student recently used Dyad to model how robots move around in space. Another used it to build a rocket engine.
“最后他说, ‘我无法’相信这有多容易—我刚刚得到了一个火箭发动机!’” Edelman回忆道。
“At the end he said, ‘I couldn’t believe how easy that was — I just got a rocket engine!’” Edelman recalls.