Data Engineer

Request for Proposals

Context 

DataKind is a global nonprofit dedicated to harnessing the power of data science and artificial intelligence (AI) in the service of humanity. Named one of Fast Company's 2017 top 10 innovative nonprofits, DataKind teams talented pro bono experts from academia and industry with visionary changemakers to collaboratively design innovative solutions to tough social challenges. We help social organizations apply their data to predictive analytics, machine learning algorithms, and AI needs in a way that both increases impact and prioritizes ethics.

See below for more information about our work and community.

Project Description

DataKind has joined forces with Jacaranda Health, an international non-profit organization that works to improve the quality of care in Kenya’s public hospitals where the majority of low-income women deliver their babies. One of Jacaranda Health’s major initiatives is its PROMPTS platform, an SMS-based service that allows mothers at all stages of pregnancy – as well as following birth – to quickly and easily receive answers to their health questions. 

In order to execute upcoming data science projects based on PROMPTS data, DataKind will construct an extract, transform, load (ETL) pipeline so that Jacaranda Health can integrate this SMS data into its newly-built, document-based data infrastructure.

DataKind is looking for a Data Engineer to scope and execute this project for Jacaranda Health by September 30, 2020

Responsibilities include:

  • Get familiar with RapidPro, the open-source mobile development platform upon which Jacaranda Health’s PROMPTS service is built, and its API.
  • Incorporate human-centered design principles to capture requirements for our project partner’s deployments.
  • Develop a project implementation plan, an audit of partner’s data, and a feasibility and risk assessment of proposed methodology(-ies), with specific considerations for future data science projects that will be completed using the partner’s data.
  • Execute on the project plan, including constructing an ETL pipeline (based in GCP) to retrieve PROMPTS data from RapidPro and load it into a MongoDB cluster on a daily basis, 
  • Write thorough documentation, and training for the partner on its operation
  • Monitor the completed deployment and troubleshoot any issues that arise prior to the end of the engagement.

Proposal

By 11:59pm EST on August 17, 2020 interested parties are requested to submit a proposal package to emily.yelverton@datakind.org that includes: 

  1. Your resume.
  2. A short description of what interests you about this opportunity and how your qualifications and previous related experience make you an excellent candidate for this position. 
    1. If you have experience working in an international context, please describe that. 
  3. A skills matrix of your expertise in using NoSQL databases, Google Cloud Platform, R/Python, and other data engineering tools where appropriate.
  4. A description of the approach you might take to scope, design and execute the project - including key milestones and estimated timeline.
  5. An estimated range of hours and rate per hour, or alternatively a specific fee for service
  6. Two references of past clients, ideally for similar work, with contacts and context. 

Please consider whether you have a nonprofit rate to offer, given DataKind’s 501(c)(3), tax exempt status. We particularly encourage women and people of color to apply.  

Selection

Strong proposals will be contacted for further discussion. These will be selected based on clarity of the proposed project plan/structure, alignment with DataKind’s goals and budget, and relevance of qualifications and past experience. 

Additional Context on DataKind

  • We are 20,000 strong: DataKind volunteers hail from almost every country across the world, representing a community of technologists and social change makers who believe in using data science and AI for a stronger planet and a more just and prosperous world.

  • We have served 300+ projects around the world: From using satellite imagery to measure crop yields to automating the fight against hate speech online, our volunteer community has shown up to help solve some of the most pressing challenges facing humanity around the world. Through this approach, illustrating what we call “the art of the possible,” these projects have resulted in more effective NGOs, more funding for social change organizations, and more awareness of the power (and pitfalls) of data science and AI in the social sector. 

  • People are at the center of our process: The most advanced machine learning algorithms are useless if they’re not designed for real people’s problems. DataKind celebrates human centered design principles that focus on the “why,” the “for whom,” and the “should we” long before we ever get going on the “what.”

  • “AI for Good” will only succeed if it’s “AI for All”: We will continue our work until all people and all communities can use digital technology for the outcomes they want to see. We support not only the tech communities of today but also anyone in any community who wants to lead how data technologies serve their needs fairly, with an emphasis on including individuals from groups that are historically underrepresented in technology or will be impacted by the intervention. 

DataKind has always known that we will not be able to combat all of the world’s looming challenges on the backs of individual, short-term projects alone. Thanks to generous long-term support from the Rockefeller Foundation and the Mastercard Center for Inclusive Growth, we are now focusing the DataKind network on issue areas, such as improving health outcomes through Frontline Health Workers or increasing economic resilience. Over a period of five years, DataKind has been granted $20M to make this vision a reality, and we are looking for people to join us in this journey. You can read more about the new strategy here. 

Apply