AI for Research & Evidence Intern

AI for Research & Evidence Intern

About IDinsight

IDinsight helps leaders combat poverty worldwide by designing, deploying and promoting evidence-generating tools. We tailor the best methodologies to partner needs and constraints to fuse evidence with action. We serve governments, NGOs, foundations and social businesses across Africa and Asia in all major program areas, including health, education, agriculture, livelihoods, finance, energy and governance.

Our diverse, growing team of roughly 250 outstanding colleagues operates in nearly two dozen countries worldwide. Learn more about our mission and values at www.IDinsight.org.

About the Role

IDinsight is seeking 1-2 summer interns to help us advance how generative AI can accelerate and enhance our data and evidence toolkit. As a research and advisory organization, we deploy a wide range of methodologies, including experimental evaluations, monitoring systems, data analytics, process evaluations, machine learning, and generative AI solutions to help partners make better decisions. We believe AI has significant potential to make this work faster, more rigorous, and more accessible.

IDinsight has had a data science and AI team since 2018. As early movers in the AI-for-good space, we now have one of the strongest AI teams in the development sector. We’re looking to accelerate how we apply this technology internally to our full suite of services. Read more about our AI strategy here, and some lessons from our internal transformation here.

This intern will work at the intersection of AI and research methodology, identifying opportunities to integrate AI tools into our evidence workflows and helping scope or pilot promising solutions.

The intern will work with Doug Johnson, Rob Sampson (both former MPA/IDs), and Anmol Rawat.

What You'll Do

Execute AI Advisory Engagements

• Conduct landscape research: Research how other organizations, academic institutions, think tanks, research firms, and development organizations, are successfully applying AI to accelerate research. Synthesize insights and identify approaches IDinsight could adapt or adopt.
• Map AI opportunities across the research lifecycle: Audit IDinsight's current research workflows (literature reviews, survey design, qualitative coding, data analysis, report writing, etc.) to identify where AI tools could add value.
• Evaluate off-the-shelf AI tools: Test and assess existing AI tools (e.g., Elicit, Consensus, NotebookLM, Claude, ChatGPT) for specific research tasks. Document capabilities, limitations, and recommendations for adoption.
• Scope bespoke solutions: Identify high-value use cases where off-the-shelf tools fall short and custom AI solutions could significantly accelerate or improve research quality. Develop requirements and feasibility assessments in collaboration with IDinsight's data science team.
• Build AI-powered tools: Develop solutions to augment research workflows, ranging from configuring AI agents (e.g., CustomGPTs, Claude Projects) for specific tasks to building more custom applications using LLM APIs, depending on your skills and the use case requirements.
• Pilot tools: Design and run small pilots to test AI-augmented research workflows, measuring impact on speed, quality, and researcher experience.
• Synthesize and share learnings: Document findings in formats that help IDinsight researchers adopt new tools and approaches. Contribute to internal knowledge sharing.

Qualifications

We're looking for a combination of research know-how, technical capability, product thinking, and entrepreneurial drive.

• Commitment to IDinsight's
  values and mission.
• Strong quantitative and analytical skills (econometrics, statistics, data analysis)
• Familiarity with research methods used in development economics and policy evaluation
• Comfort with statistical software (R, Stata, or Python)
• Curiosity about how AI is changing knowledge work and research
• Strong written communication skills
• Experience using off-the-shelf generative AI tools (ChatGPT, Claude, etc.) for substantive work
• Nice-to-have: Experience building custom AI applications using LLM APIs.

Nuts & bolts

Location: We prefer for this person to be based in New Delhi, to overlap with Anmol and other AI and Research work, though we are open to the person being based in one of our other offices (Rabat, Dakar, Lusaka, Nairobi, Manila).

Start date: Flexible

Duration: 10 weeks minimum, priority given to candidates who can commit to 12 weeks.