Meet Algo, your new boss. It’s flexible, willing to change work schedules so you can work when you want, and not when you don’t. It’s reasonable, providing you honest feedback without the politics of your last human boss. And Algo will find new projects for you to complete, so you always have something interesting to work on while receiving a paycheck. The best part is, “Algo” is pretty much here. We all know the story about algorithms and work the past few years. Service jobs across the country are increasingly being managed with the help of mathematical models of customer demand, revolutionizing everything from taxi driving to food delivery, home cleaning, and laundromats. I have argued that the increased autonomy and flexibility of these jobs means that algorithms are taking over unions as the primary driver of workers’ rights in the 21st century. But now, startups are starting to move up the corporate ladder, using algorithms to improve and disrupt professions that up until recently have seemed almost completely insulated from the efficiencies of computation. Two Boston-based startups, HourlyNerd and Quantopian, are using sophisticated algorithms to compete with some of the most-well known firms in consulting and quantitative finance, respectively. As they begin to disrupt these two industries, both hope to expand and democratize their professions to anyone with the talent and desire to work. If they are successful in their ambitions, we may start to see a world where elite and inaccessible credentials are replaced by performance-based reviews, creating a much more meritocratic workforce than we have today. It will also provide workers with far more flexibility in their daily lives, something that is heavily demanded by today’s workforce. And while flexibility does often come at the cost of job security, this new wave of startups promises to provide great livelihoods to a wider number of people. Consulting Needs A New Consultant There is a running joke about consultants: they charge you a lot of money to tell you what you already know. While top management consulting firms may emphasize their unique skillsets and talents to prospective clients, the reality is that the vast majority of consulting work falls into a handful of routine project types like competitive analysis, new market entry, and product pricing that most MBAs are trained in business school to perform. HourlyNerd is hoping to open up the consulting profession by connecting potential clients through a marketplace to their large number of independent consultants. They have already seen significant traction, with several five-figure projects completed and several larger ones in the works. It may seem weird that your human boss could be replaced by a computer, but that unfamiliarity is likely to be soothed by the freedom of the work that you will be able to do in the future. The company was founded in January 2013 by Rob Biederman and Pat Petitti, two MBA students from Harvard Business School who thought that the current model of consulting just didn’t make sense in a world of flexible workplaces and democratized marketplaces. Based on their own experiences in management consulting and investment banking, the founders designed a product for companies looking to hire top consulting talent at more wholesale rates than full-service firms provide. The company caught the eye of investors, garnering $4.8 million in capital from the likes of Highland and Greylock earlier this year. Traditionally, a group of consultants, led by an engagement manager and backed up by a coterie of associates and analysts, decamp to a client’s corporate headquarters for 3–6 months to conduct interviews, collect data, and perform analysis to come up with a solution to a defined business problem. These consulting deals are often worth millions of dollars, preventing all but the richest companies from getting access to top talent. At the same time, the consulting team gets worn out from the grueling weekly travel schedule. HourlyNerd sees itself as both more responsive and more valuable to companies. The startup guides perspective clients through building a project proposal and selecting talent, using algorithms and previous performance data to find the right match. Unlike consulting firms, which may take weeks or even months to get a team in place, HourlyNerd may be able to connect a company to a consultant in just days, accelerating the business cycle for managers. On the other side of the marketplace, the company recruits MBAs from top schools, with the hopes of expanding the talent pool as it learns more about the qualities that make for the best consultants. When they first start, new consultants are put on relatively simple projects to gauge their abilities. As they prove themselves to clients, they transition to projects that are more complex. For consultants, the top benefit of the platform is the flexibility. They can travel the world while earning money a few weeks at a time, or they can work full-time, but only on projects of their choosing. That level of control is unique in the industry, and a key selling point to entrepreneurial business students who otherwise have valuable competing offers to consider. Using Algorithms In Algorithmic Finance Quantopian has a simple vision for the future of algorithmic finance: at home, in shorts and a t-shirt. The startup is developing an algorithmic trading platform that will allow quants from anywhere to craft their own investment strategies and potentially strike it rich. When TechCrunch first discussed the company last year, the startup was still building out its trading platform and algorithm design tools, and ignoring how it was going to make money. Now, with its platform fully formed and flush with a brand new round of capital from Bessemer to the tune of $15 million, the company is preparing for an even more audacious goal: the first democratized hedge fund in the country. The company, led by CEO John Fawcett, is hoping to eventually open its platform to all investors so that anyone with the right performance track record will be able to manage a part of a central pool of capital – and profit from the returns. That’s a massive change, since today’s hedge fund managers can’t just be excellent quants, but must sell their ideas to prospective limited partners, a rarefied world that is mostly inaccessible to all but bankers on Wall Street. Quantopian will constantly monitor the strategies of all the algorithmic traders on its platform, identifying those with the best returns with minimal risk. It will then build and adjust its portfolio to carefully select a balanced set of these strategies. Instead of the one-hit wonder that plagues the hedge fund industry today, the startup believes that it can create sustainable growth as it moves from one effective algorithm to another. The challenges to its model though are legion. As Fawcett explained to me recently, building such an algorithm is extraordinarily difficult, since there are real challenges in balancing how algorithmic traders use their own accounts with how they might trade with others’ capital. There are also complications arising from the diversity of algorithms that might be used on the platform to the risk models available to Quantopian to judge them. These issues are vexing, but not intractable. Algorithms And The Workplace It may seem weird that your human boss could be replaced by a computer, but that unfamiliarity is likely to be soothed by the freedom of the work that you will be able to do in the future. For the first time, independent professionals may have better access to interesting projects than even the best firms in their industry. They will have a global network of opportunities to select from, which not only makes work more interesting, but also brings new challenges that will allow these professionals to develop their careers more rapidly. Perhaps even more importantly, professionals will be able to cross marketplaces, breaking the categorization that is endemic in these fields. A doctor who wants to do business consulting on the side will be able to do so, improving their practice. Rather than putting workers in boxes, this new wave of startups has the ability to usher in a unique period of creativity for talented people. And we are likely to see more of these startups. Zavain Dar, who recently joined Lux Capital as an investor, believes that algorithms are hitting their stride due to a confluence of three factors. First, we finally have the data infrastructure to be able to process lots of data, using technologies like Hadoop. Second, we now have the ability to digitize massive data into unstructured datasets, often stored in the cloud. Finally, we have the ability to analyze this unstructured data in intelligent ways that were impossible just a few years ago. That’s in part why we are seeing this move up the corporate ladder by startups. Up until now, most talent marketplaces have been based around demand forecast models, in which workers are matched with clients as needed, whether that means getting a vehicle to a certain part of a city or a meal to a home. While certainly not a trivial task, these algorithms are also common enough to be readily implementable. But what about managing a hedge fund portfolio or identifying the best consultant for a project? Suddenly, there is much more nuance involved, and the algorithms will have to become equally sophisticated. Dar emphasizes the product feedback loop. “From a venture perspective, it is surprisingly easy to be an entrepreneur and look at a cool dataset and a cool algorithm and have the infrastructure to use it, so that you can go and build a startup. I don’t think that is going to be a long-term company.” Instead, he emphasizes that entrepreneurs should build products that can use data to make the algorithm better over time. “Everyone knows how Google’s algorithm works, but no one has the same data that Google has to train that algorithm and make it work really well. They have probably the most proprietary and valuable dataset, which is the clickstream data.” That is perhaps the best point of these sorts of elite talent startups. As more people and projects join them, the quality of their algorithms will increase, creating a positive cycle of improvement. Yes, There Are Dangers Of course, there are real concerns as we move away from traditional models of employment to more flexible ones. Job security is conditioned on performance, but what happens if a client is unhappy and tries to end your career by destroying your credibility? Startups will have to balance the desire of clients wanting past performance data with the need to recognize how talent develops over time, so that an early mistake in a career isn’t deadly to a worker’s future prospects. Coupled with performance is simply the issue of salary security. If you don’t work, you don’t get paid. That means that parents who would like to take time off to be with their newborn, or a worker who gets sick and needs to take leave will be able to do so, but at a direct financial cost. There are likely ways to mitigate such impacts, but these are real challenges to solve. Beyond individuals, automation and efficiency have already eliminated entire professions and downsized others. Companies no longer account for their profits using rows of humans crunching numbers, and telephone operators are no longer a typical occurrence when we dial long-distance. Up until now, automation has been far less common in white-collar and elite professions like medicine, law, banking, and consulting. That is changing, and fast. Ten years ago, the market for lawyers in the United States was healthy, taking in thousands of new legal graduates a year with salaries far more than double the median income in the United States. Now, the market has cratered, with many graduates reporting that the best jobs they can find are paying $12–15 per hour. Dozens of legal startups like RocketLawyer and LegalZoom offer services that used to cost thousands of dollars for just a fraction of that price. So far, the fate of lawyers hasn’t yet befallen management consultants or bankers. But what happens when work becomes cheaper in these professions as well? Ironically, the great income inequality that we have witnessed over the past 30 years may end up becoming even more acute. Ultimately, our best bet may be a market that is a combination of the old and the new, perhaps driven by algorithms in the middle helping us match work to workers. Such a hybrid market will allow us to choose between extreme flexibility and no security, or security with limited flexibility. For the first time, we will have a say in the matter, and all we have to do is ask Algo.