Machine Learning Scientist: Forecasting
Build machine learning systems for environments deemed unforecastable: predict where and when medicine, vaccines, equipment, human resources and supplies are needed.

Macro-Eyes is recruiting a Machine Learning Scientist with a focus on forecasting to build cutting-edge machine learning systems that power the first self-driving supply chains. You will create products that automate the matching of supply to demand in mission-critical settings where there is extreme uncertainty about both supply and demand.

The solutions you build will provide companies and governments radical improvement [order of magnitude] in the capability to predict demand, risk and opportunity across crucial supply chains.

You will create transformative solutions the world has never seen.

We are working on unsolved problems at the nexus of state-of-the-art of machine learning and crucial supply chains: the first predictive supply chain for health; a new, deployed system for understanding demand for lifesaving care; computer vision and deep learning to automate ground-truth data collection when compute has to run entirely on-device; deep learning on satellite imagery to derive features that help us understand the ‘context for care;’location inference accurate to within 25M when GPS is unreliable; human-in-the-loop machine learning with experts on the frontlines to learn what we don’t know and generate real-time insight.

Go beyond theory: what you build will become products, deployed across the developing world — and for some of the most sophisticated consumers of technology in the world — to shift crucial supply chains from reactive to proactive, capable of anticipating risk and opportunity for governments and corporations. Your work will save lives.

The Macro-Eyes team is spread across Africa, Europe, and North America. We believe in recruiting the best talent in the world, regardless of location. This role will be fully remote with the freedom to choose how and when to work. Our remote team was executing at the highest level before COVID-19, and we expect to continue to be a fully remote, high-performing team. You will have access to your choice of hardware and a travel budget to interact with the distributed team in person.

  • Work with statistics and unconventional data to drive the use of ML for social good.
  • Building models based on empirical results, and understanding of the strengths and limitations of this approach.
  • Build machine learning systems for forecasting health commodities.
  • Make meaningful contributions to projects with a research flavour.
  • Be aware and provide current best practices in machine learning.
  • Experience working with diverse data types: images, structured data and natural language.
  • Experience programming in Python is a must, and one additional language (R, C, C++, Java).
  • Knowledge of statistics, including hypothesis testing with parametric and non-parametric tests and basic probability.
  • Experience building different deep neural network models or knowledge of graph theory.
  • Demonstrated ability to develop and implement creative approaches for making machine learning work with limited conventional data.

We’re excited if you also have these capabilities:

  • Experience working with teams using Agile development processes.
  • Experience with international donor-funded projects.

What we’ll ask of you:

  • Take a task and run with it. There’s not a lot of hierarchy, and everybody organically occupies the space they’re given to get something done.
  • Speak up. We all learn every day. We constantly figure out things we’ve never done before. Ask for help, say when you need something.
  • Be forthcoming and accountable. About time, scope, questions, needs, options, deliverables.
  • Be nice to work with. Nothing extraordinary was ever achieved alone.

What you need to bring:

  • A data-first mindset with a proven track record of delivering successful products.
  • Comfortable working with a small team in a fast-paced, collaborative but unstructured environment.
  • Experience working on dynamic cross-functional teams.
  • Strong written and oral communication skills.
  • Energy, boundless curiosity, open-mindedness.
  • Care to do something right for the sake of doing it right.
  • Willingness to learn and grow.
  • Respect for audiences and users. Willingness to step into someone else’s shoes (and the imagination to do so).

Why you want to work with us:

  • We develop technology and products that save lives and help organizations do more with fewer resources — and we break new ground in AI by creating systems that can learn in some of the most challenging environments on earth: real-world impact and state-of-the-art science.
  • We are rapidly growing — globally — and engaged in many of the most important issues of our time. Being part of Macro-Eyes is rich in opportunity.
  • You get the chance to shape the world, bend it towards progress!
  • You get to work with an amazing group of people.
  • You can work from almost anywhere, although less than 6 hours from EST tends to be optimal.

Competitive compensation and generous benefits.

Macro-Eyes has a rigorously horizontal culture that values diversity of every kind, and we are dedicated to building an inclusive workforce where diversity is valued. Macro-Eyes is an equal opportunity employer. Every qualified applicant will be considered for employment. Macro-Eyes does not discriminate based on race, color, religion, gender, gender identity or orientation, genetic information, age, national origin, marital status, disability status, political ideology, military or protected veteran status, or any other characteristic protected by applicable federal, state, or local law.

Disclaimer: Due to large volumes of applicants, only shortlisted candidates will be contacted.

Warning: the work will not be like doing data science at Big Tech: the data will not be pristine and what’s required for success is much harder than throwing a series of models at the problem. We expect you to think in terms of hypotheses and experimental design like a scientist and have the drive and ingenuity of a hacker with hands-on experience building predictive models.