Director of Data & AI

About Think of Us

Think of Us (TOU) is transforming the child welfare system into one where young people and families can heal, develop, and thrive. We drive purpose-driven work aligned with meaningful systems change, powered by data and the insights of lived experience. Our focus is on prevention, keeping families strong before crisis, and on building kinship-first systems so children grow up with people they know and trust. Through bold collaborations with leaders and innovators across the public and private sectors — from Presidential Administrations to local providers — we identify leverage points and co-create solutions that re-architect child welfare for the future.

Here’s what this looks like:

Built the Lived Experience Engine into the largest dataset of its kind (51,000+ voices), powering AI tools and shaping reforms

Partnered with California to launch the Kinship Accelerator in 8 counties, backed by $150M and a budget increase from $317M → $896M

Drove a national kin-first shift: 39 states (plus 5 tribes) adopting kinship standards, unlocking $3B for families

Operate one of the largest direct resource networks, connecting 1,827 families to $28M+ in support

Our team members share a commitment to driving systems change through creative problem-solving, obsessive curiosity, and a love of quickly driving innovative, measurable, and impactful results.

Position Overview

Department: Technology & Innovation

Reports to: Executive Leadership

Direct Reports: Data Engineer, Senior Applied Data Scientist

Stack: Python, SQL, Databricks (bronze/silver/gold). AI pipeline on Anthropic Claude +

OpenAI, with LangFuse for observability and evaluation.

Location: Hybrid; Washington, DC (min. 3 days in the office)

Category: Full Time/Exempt

As the senior leader for data at Think of Us, you'll be a working player-coach: contributing hands-on alongside the existing data team for the first 6–12 months before growing it further. You're not a pure strategist or a pure manager: you've built a data function before and know how to do the work while building the systems around it.

Data is one of TOU's most powerful strategic assets, and none is more central to our mission than the largest dataset of its kind: perspectives from over 51,000 individuals impacted by the child welfare system. As TOU integrates AI across its operations, expands geographically, and deepens its policy and programmatic impact, we need a leader who can turn this proprietary dataset into a durable technical advantage. You'll lead TOU's data function end to end: strategy and architecture, AI and evaluation, analysis and insight, and governance. You'll directly manage the team that delivers this work, today a data engineer and applied data scientist, and grow it over time.

Your data team also powers TOU's applications, most centrally ConnectMe, our warmline platform where Care Navigators work directly with youth and families. Its AI pipeline drafts assessments, support plans, and resource matches, always reviewed and owned by a Community Responder before anything reaches a help-seeker. Your team owns the evaluation layer and quality bar that tell us whether any of it is actually working, all within a HIPAA-compliant environment.

You'll partner closely with executive leadership on technical direction, with the engineering team on AI infrastructure and quality, and with Product, Design, Program, and Policy leaders on how data informs what we build, ship, and advocate for. We operate as a modern data team: framing the right questions, shipping insights and evaluations into product and policy decisions, and iterating based on what we learn from real-world data.

Key Responsibilities

Own Data Strategy and Architecture

Define and drive TOU's data strategy and roadmap, anchored on the Lived Experience Engine

Direct the architecture of our Databricks data lake: what goes in each tier, how the data model evolves, and what standards apply

Set standards for data quality, governance, and platform reliability

Translate organizational priorities into data investments and team structure

Lead AI Evaluation and Applied Data Science Direction

Build evaluation frameworks (LangFuse plus custom) for TOU's AI pipelines and agents, the systems that tell us whether AI is actually working

Own decisions on when and how TOU goes beyond cloud models: RAG architecture, agent grounding, retrieval quality, and future ML capability

Negotiate the boundary between the data team and Applied AI Engineering on evals and AI data infrastructure

The data team owns the data and evaluation layer that makes agents work; Applied AI Engineers own the engineering of agents themselves

Own Governance, Privacy, and HIPAA Compliance

Build and maintain TOU's data governance posture and HIPAA-adjacent compliance standards

Own data access controls, privacy policies, and the ethical framework for lived-experience data

Ensure TOU's handling of sensitive PII from youth in the child welfare system meets legal and ethical standards

Build and Manage the Data Team

Directly manage the data team (currently a Sr. Data Engineer and Sr. Applied Data Scientist), and grow the team over time

Build the processes, culture, and norms of TOU's data function

Be IC-first for 6–12 months, with management load growing as the team grows

On a Typical Day, You Might…

Give architectural feedback on a new Databricks gold-layer model the Data Engineer is building

Design an evaluation framework for the ConnectMe AI pipeline, defining what "good enough" means for human-vetted output

Meet with program leadership to translate research questions into data roadmap priorities

Review AI pipeline outputs to assess whether resource recommendations are accurate and equitable across demographic groups

Sit with the Sr. Applied Data Scientist on a recent eval, then decide what's ready to ship, what needs another iteration, and what to brief Product on

Draft a data access policy for how lived-experience narratives are stored and accessed internally

About You

We recognize that candidates may not meet every listed qualification. Research shows that individuals from underrepresented groups are more likely to opt out of applying if they don’t meet 100% of the criteria. If you believe you meet many of the criteria and can succeed in this role, we encourage you to apply — we want to learn about your unique strengths and experiences.

Qualifications:

7+ years in data, analytics, or ML roles, with at least 2 years in a leadership or team-building capacity

Direct, hands-on experience with applied AI such as RAG, agents, and LLM evaluation; candidates without this will not be considered

Technically credible: can design data architecture, read and review Python, evaluate AI pipeline quality, and hold substantive technical conversations with engineers

Has built or led a small data team before, in the 2–8 person range; what matters is that you've done it, not the scale

You bring experience working with or alongside systems that impact youth and families — or you’re deeply committed to learning about these systems and approaching the work with humility, curiosity, and care.

You’re motivated by impact and systems change. You thrive in a fast-growing, evolving organization and see change as an opportunity to innovate, learn, and make things better.

Desired / Nice to Have:

Experience with Databricks, Delta Lake, or similar modern data lake platforms

Experience with LLM observability and evaluation platforms (LangFuse or similar)

Background in civic tech, public health, human services, or mission-driven organizations

Experience with data governance or HIPAA-adjacent compliance

Prior experience as a player-coach who remained hands-on while building a team

How You Work:

Lead as a player-coach: you are as comfortable doing the work and building the systems as you are growing the peop