Consultancy - Research Scientist, Machine Learning Specialist, Office of Innovation

UNICEF Office of Innovation is looking for a research scientist, with a strong focus on optimization, to join our Science Team remotely, part-time for 6 months (equivalent to 3 months full time). The consultant will help move forward our work on analysis of multiple datasets with machine learning algorithms. We are looking for motivated and experienced researchers in data science related fields (computer science, applied mathematics or physics) and a deep interest in applying research towards humanitarian and development work.

UNICEF works in 190 countries and territories to protect the rights of every child. UNICEF has spent 70 years working to improve the lives of children and their families. Defending children's rights throughout their lives requires a global presence, aiming to produce results and understand their effects. UNICEF believes all children have a right to survive, thrive and fulfill their potential – to the benefit of a better world.

Duration: 6 months half-time (up to 63 working days)

Duty Station: NYHQ

Level: Individual contract (equivalent to P3)

Background:

UNICEF Office of Innovation is looking for a research scientist, with a strong focus on optimization, to join our Science Team remotely, part-time for 6 months (equivalent to 3 months full time). The consultant will help move forward our work on analysis of multiple datasets with machine learning algorithms. We are looking for motivated and experienced researchers in data science related fields (computer science, applied mathematics or physics) and a deep interest in applying research towards humanitarian and development work.

UNICEF Office of Innovation combines several disciplines such as network science, natural language processing, data mining, machine learning, remote sensing, artificial intelligence or complex systems in research lines such as human dynamics, epidemiology, urbanization, migration or influence dynamics.

Candidates should expect to pursue research leading to both high level scientific publications and operational tools for UNICEF. Current lines of work range from epidemic modelling and response in emergencies such as Ebola and Zika, understanding urban dynamics and migration resilience in crisis such as the Syrian refugees, preparedness and early warning for natural hazards such as El Nino or understanding how children rights get embedded into legal and corporate policies. Current data sources include mobility and communication networks, social media and news, satellite images, financial and economic data, weather and environmental data along with a wide variety of traditional humanitarian and development indicators.

The Science Team interacts with world class scientists, on the ground responders, private sector data holders and tech leaders to push the Data Science agenda towards the problems that affect the most vulnerable children. Breakthroughs and working methodologies are streamlined to a dedicated Development Team to scale into operational tools that can make a real impact.

UNICEF Office of Innovation functions like a collection of startups in early-stage in which promising new technologies (e.g. data science, SMS technologies, UAVs, blockchain) are identified that have potential to impact UNICEF core operations (e.g. education, health, disaster response). Through our network of technology partners, UNICEF country offices and academic institutions, we construct pilots to quickly determine potential impact. The Office of Innovation provides support to UNICEF's 135 country offices and consists of a flat team of technologists, designers, programmers, researchers and communication specialists. Requests for collaboration, advice and support are fielded from UNICEF programmes, other parts of the UN system and the humanitarian technology ecosystem.

Purpose of the appointment:

The Office of Innovation at UNICEF is in need of a machine learning specialist familiar with cutting edge machine learning algorithms (e.g. deep learning, transfer learning) and their applications on image analysis. The consultant will be working with other data scientists in the office of innovation team to develop and apply machine learning algorithms to produce socio-economic indices such as poverty index from satellite data. In our team, we work on various development related topics using big data, satellite imagery and other types of geospatial data. As a critical part of such research endeavor, the consultant will work to train machine learning models out of the available data and apply trained models on satellite data to extract indices or classify satellite imagery into thematic maps.

Payment Plan:

• Monthly payment, based on monthly tasks and progress reports, approved upon monthly review with supervisor.
• Consultant is responsible for his own health insurance
• Consultant is eligible for standard DSA for all work-related travel.

Qualifications and Education:

• University degree in computer science, geospatial science, statistics or comparable academic preparation.
• Minimum 5 years of experience in machine learning applications.
• Proven familiarity with machine learning algorithms and its application on remote sensing data.
• Excellent knowledge of English is required.

Specialized Experience and Knowledge required for the assignment

• Familiarity with geospatial data and processing using open source library such as GDAL.
• Programming ability in Python or R.
• Strong communication skills (written and oral) are a necessity, and proven analytical skills in conceptualizing and developing literature reviews. Strong visual communication experience is preferred.

Languages

Fluency in English.

Applications must include:

• CV,
• 2 selected scientific publications of the applicant's portfolio with a small description of why they were selected for this application
• Link to google scholar profile, github profile, medium or blog site (if available)

UNICEF is committed to diversity and inclusion within its workforce, and encourages qualified candidates from all backgrounds to apply.