SXSW Startups: Molecula
The Forrest Four-Cast: February 25, 2019
Fifty diverse startups will aim to impress a panel of judges and a live audience with their skills, creativity and innovation at SXSW Pitch Presented by Cyndx. Winners in 10 categories will be announced at the Pitch Award Ceremony at 6:30 pm Sunday, March 10, at the Hilton Austin.
A finalist in the Enterprise and Smart Data category, which will pitch at 12:30 pm Saturday, March 9, Molecula is a continuous analysis platform based on Pilosa. Their superpower is joining data lakes with data streams to power real-time, intelligent applications. Their mission is to allow scientists and engineers to focus on the fundamental research needed to solve complex problems without the expensive and difficult data management usually required.
CEO Higinio O. Maycotte answered some questions about his Austin-based company.
What do you like most about the startup experience?
The people. I love nothing more than bringing great people together to try to do the unimaginable. One of the best feelings in the world is inspiring the best in an individual or team and seeing them achieve something that has previously been deemed impossible.
What’s one piece of advice you wish you’d had that you’d give to others wanting to join the startup journey?
It is 100 percent about the people you surround yourself with. Focus on A++ humans personally and professionally. This includes co-founders, management, team member, board members, investors, clients, etc. The second you make compromises at any level, you bring your whole company down to the lowest common denominator.
You now run Pilosa, what you describe as “an open source, distributed index designed to help analyze huge datasets at UI latency.” How will it work with Molecula?
Pilosa is a discrete technology that makes queries on large data extremely fast. However valuable it can be, the conceptual and technological adoption curve for Pilosa can be daunting for even the most sophisticated engineering teams. To solve this, Molecula is turning Pilosa into an easily adoptable managed service. We are also focused on our primary data consumer being a machine and not a human. We aim to become the standard data supply chain into machine learning and AI pipelines. Because of significant performance benefits, we are initially supporting Oracle’s Cloud OCI, launching Molecula on March 26 at Oracle’s Open World Singapore. We expect to roll out other clouds throughout the year.
What are your goals for Molecula in 2019?
When we launched our company two years ago, we open-sourced our technology and now have 1,600 companies using it. This year we are wrapping our core technology, Pilosa, into the managed service called Molecula to make accessing the largest data lakes and the most voluminous data streams as easy and fast as managing files in dropbox. We are renaming the company from Pilosa to Molecula too. While we can effectively sell our core value proposition today, it will take us the remainder of the year to fully automate our offering. In the meantime we are narrowing our focus to three verticals (bioinformatics, connected devices, security/fraud) and are choosing our clients very carefully to tell the best success stories we can. We are also in the process of building a world class management team and a distributed engineering team of the best developers no matter what time zone they reside in. Finally, our primary financial goal is to triple sales this year vs. last.
What is your competitive advantage?
Our unique advantage is that most data retrieval technologies make copies of your data to organize it in ways that speed up queries and instead we use Pilosa to facilitate access without making any copies of the data. Pilosa turns data into what is called a bigraph, also known as a Knowledge Representation, which simply maps intersections between data that are a fraction of the size of the original values, but retain all the query-able capabilities of the data. This approach practically eliminates the need to have a database, a data warehouse, a batch layer or a speed layer. More importantly, it eliminates the need to have to make the usual compromises IT teams make to accelerate access to data including caching, batching, sampling, pre-computing, etc.
Tell us something about your previous experience at SXSW.
SXSW is the most amazing concentration of serendipitous connections in the world, and it only happens once a year. Exactly what will come of each SXSW is unknown, but it is always of immense value and the key is to simply put yourself out there. Go to as many events as you can and be open to meeting new people and having new experiences. At my previous company, we presented at SXSW Pitch six years ago and the visibility we received was incomparable. We met investors, employees, clients, media and just about every valuable connection we could imagine.
If your team members weren’t involved in building Molecula, what would they be doing?
We have all worked together at two, three and in some cases four startups, so we would likely be working on another startup. If we were to go start something new today, we would probably focus on this new breed of real-time, intelligent applications we are calling “co-pilots.” These are far beyond a simple digital assistant. We are seeing companies build co-pilots that are helping us make better decisions as gamers, parents, employers, paramedics, etc. by augmenting our memory, allowing us to pool our memories and predict outcomes.
Which tech trend is your team most excited about?
We get excited about anything that unlocks the power of data in an organization. As the world has moved from storing raw data on hard drives in 1956, to organizing that data in databases in the late 1970s to the Hadoop driven data warehouse era of the last decade, we have ended up with data infrastructures that resemble very fragile patchworks of interconnections where only 20 percent of the data is even analyzable. A company called Confluent, who spun out of LinkedIn, has commercialized an open source technology called Kafka. Kafka, otherwise known as a data pipeline, is driving the most profound and valuable change to data infrastructures since we started storing data digitally. Kafka turns these patchworks into a simple architecture where all data moves through a central bus and data is either being created (by a producer) or consumed (by a consumer). This brilliantly simple approach brings an organizational agility that will usher in the Intelligence Era, where machines are going to need access to 100 percent of the data, instantly.
Look for more interviews with other SXSW Pitch finalists in this space between now and March.
Click here to see all 50 finalists for SXSW Pitch 2019, along with the links to their interviews on Medium.
Also, if you are an entrepreneur, check out all the cool panels and presentations in the Entrepreneurship and Startups Track, which runs March 8–12 at SXSW.
Hugh Forrest serves as Chief Programming Officer at SXSW, the world’s most unique gathering of creative professionals. He also tries to write at least four paragraphs per day on Medium. These posts often cover tech-related trends; other times they focus on books, pop culture, sports and other current events.
