Senior Engineer, Digital Products
The Senior Engineer, Membership Center & Ask Aya is the primary engineer on Ask Aya and a core engineer on the Membership Center platform that houses it. Ask Aya is a digital AI tool built for domestic workers with exciting early Beta results and 2,000 current users. You will lead iterations, feature development, and integrations that make Ask Aya more useful, safer, and more measurable in its impact on domestic workers' lives—working in close partnership with product, design, data, and organizing teams.
This role is a temporary role running from July 20, 2026 through February 2027, with the possibility to convert to a regular at-will position based on performance, funding, and organizational needs. The initial term is scoped so that the right person can land, ship meaningful improvements to Ask Aya, and establish the analytical foundation described below.
This is a senior hands-on role for an engineer who is deeply fluent in modern AI/LLM application development and who writes production code every day.Working alongside the Senior Director of Product and a contracted Senior Engineer who developed the current architecture, integrations, etc.You will own the architecture and delivery of LLM-powered features—prompt design, retrieval, evaluation, safety, analytics pipelines, and the integrations that connect Ask Aya to the rest of our stack (Membership Center, CRM, SMS, web).
You will also lead development on features and integrations that can aid in distribution of Ask Aya.You will also shape how we learn from Ask Aya over time: designing privacy-preserving analytical pipelines that turn worker conversations into structured insight about knowledge, attitudes, and behaviors—without compromising trust.
The ideal candidate combines strong engineering fundamentals with real fluency in AI-assisted development: someone who uses Claude Code and similar tools as a daily force multiplier, who has shipped LLM features to production, and who cares about building technology with and for low-income workers, immigrants, and movement organizations.
NDWA is working on the frontlines of AI development for good with many key nonprofit and Frontier organizations. This role is an exciting opportunity for an engineer who has frontend and backend experience, experimentation chops, and a massive interest in AI development, evaluation, and is ready to be a nonprofit champion for using AI to expand impact.
Primary Roles & Responsibilities
Ask Aya — Primary Engineering Ownership
- Serve as the primary engineer on Ask Aya, leading iterations, feature development, and integrations.
- Own the full LLM application stack: prompt design and versioning, offline evaluation suite, retrieval (RAG) over NDWA's knowledge base, tool/function calling, multi-turn conversation state, content moderation, and fallbacks.
- Ship features that measurably improve Ask Aya's helpfulness for domestic workers—e.g., drafting messages to employers, explaining legal entitlements, surfacing benefits eligibility, and connecting workers to local NDWA chapters and campaigns.
- Build and maintain a rigorous evaluation harness: regression tests on prompts, eval sets grounded in real worker needs, working with product team on red-teaming for safety, and bilingual (English/Spanish at minimum) quality checks learning implementation.
- Continuously improve latency, cost, and reliability of the Ask Aya runtime as well as guardrails and product features that drive engagement.
Membership Center Platform
- Contribute as a senior engineer on the broader Membership Center platform—the web/app experience where domestic workers sign up, access resources, and engage with NDWA.
- Build and maintain integrations between Membership Center, Ask Aya, and NDWA's organizing stack (EveryAction Email & CRM, SMS platforms: Mobile Commons, webpage, data warehouse, and Mobilize integration).
- Ensure authentication, accessibility, internationalization (especially Spanish), and mobile-first performance for users with limited bandwidth and older devices.
AI-Assisted Engineering Practice
- Use AI coding tools as a core part of your daily workflow—for scaffolding, refactoring, test generation, etc.
- Establish patterns, prompts, and guardrails so the rest of the team can safely and productively use AI tools in their own work.
- Stay current on the rapidly evolving AI/LLM landscape and translate relevant advances into pragmatic improvements in our stack and in our product infrastructure.
Collaboration, Craft & Culture
- Work closely with product managers, designers, data, organizers, and policy staff to turn worker needs into shippable features.
- Write clear technical specs, RFCs, and runbooks; review code with care; mentor more junior engineers and oversee contractors.
- Be a thoughtful voice on privacy, safety, and ethics—particularly as they affect immigrant workers and workers in vulnerable labor arrangements.
Qualifications
Required
- Senior-level engineering experience — 7+ years building and shipping production web/product software, with meaningful time owning systems end-to-end.
- Hands-on LLM application experience — has designed, built, and shipped LLM-powered features in production (not just prototypes). Comfortable with prompt engineering, RAG, tool/function calling, evaluation, and handling non-determinism in real systems.
- Deep understanding of AI fundamentals — can reason about model selection, context windows, cost/latency tradeoffs, hallucination and safety failure modes, structured output, and eval design.
- Fluent use of AI coding tools — uses Claude Code or equivalent daily; can describe concrete ways AI tools have changed how they ship, and can coach others to do the same.
- Strong full-stack fundamentals — TypeScript/JavaScript, Node.js, React/Next.js, REST/GraphQL, SQL.
- Data layer experience — Prisma (or comparable ORM), PostgreSQL, schema design, migrations.
- Cloud & infrastructure — AWS (Lambda, S3, RDS, ECS/Fargate or similar), CI/CD, observability, and basic infrastructure-as-code.
- Data pipelines & scheduled jobs — has built ETL or analysis pipelines (cron/EventBridge/Step Functions/Airflow-class tools), with proper retry, idempotency, and monitoring.
- LLM analytics & structured extraction — experience using LLMs to analyze unstructured content (e.g., conversations, documents) and produce durable structured outputs—topics, signals, flags, intent—that become a permanent analytical layer downstream systems can rely on.
- Framework-driven conversation analysis — comfortable designing LLM-graded prompts against a research framework (e.g., KAB—Knowledge, Attitudes, Behaviors) to evaluate conversations for things like whether a user received relevant information, showed shifts in confidence, or surfaced intent signals such as asking for help drafting a message to an employer.
- Privacy-preserving data architecture — can design end-to-end pipelines that run LLM analysis within our own infrastructure under a ZDR (Zero Data Retention) agreement, write structured outputs back to a Prisma/AWS setup, and delete source content on a defined retention schedule.
- Privacy & security mindset — experience handling sensitive user data, working within retention and -style constraints, and designing systems that minimize collection.
- Clear written communication — can write specs, decision docs, and runbooks that other humans actually use.
Strongly Preferred
- Experience with the Anthropic/Claude API — including streaming, tool use, prompt caching, and working under a ZDR agreement.
- Evaluation & analytics for LLM systems — has built eval harnesses, offline grading pipelines, or LLM-as-judge systems for quality and safety.
- Spanish language skills — either conversational Spanish or experience and comfortability building bilingual (EN/ES) product experiences.
- Mission experience — prior work in civic tech, campaigns, worker or immigrant advocacy, public benefits, or other movement/organizing contexts; or experience building technology for and with low-income communities.
- Integrations with organizing stacks — EveryAction, Mobilize, SMS platforms (broadcast and P2P), or similar.
- Accessibility & internationalization — shipped products that meet WCAG and work well on low-end mobile devices and slow networks.
What Success Looks Like in the Initial Term (June – December 2026)
- Ask Aya measurably improves on helpfulness, engagement, accuracy, and safety, with a durable evaluation system behind every change.
- Online and Offline evaluations are dependable for Ask Aya.
- Engineering requirements tied to distribution and reengagement and adoption hypothesis are scoped out through December.
- Sprint planning accurately reflects
Senior Engineer, Digital Products
The Senior Engineer, Membership Center & Ask Aya is the primary engineer on Ask Aya and a core engineer on the Membership Center platform that houses it. Ask Aya is a digital AI tool built for domestic workers with exciting early Beta results and 2,000 current users. You will lead iterations, feature development, and integrations that make Ask Aya more useful, safer, and more measurable in its impact on domestic workers' lives—working in close partnership with product, design, data, and organizing teams.
This role is a temporary role running from July 20, 2026 through February 2027, with the possibility to convert to a regular at-will position based on performance, funding, and organizational needs. The initial term is scoped so that the right person can land, ship meaningful improvements to Ask Aya, and establish the analytical foundation described below.
This is a senior hands-on role for an engineer who is deeply fluent in modern AI/LLM application development and who writes production code every day.Working alongside the Senior Director of Product and a contracted Senior Engineer who developed the current architecture, integrations, etc.You will own the architecture and delivery of LLM-powered features—prompt design, retrieval, evaluation, safety, analytics pipelines, and the integrations that connect Ask Aya to the rest of our stack (Membership Center, CRM, SMS, web).
You will also lead development on features and integrations that can aid in distribution of Ask Aya.You will also shape how we learn from Ask Aya over time: designing privacy-preserving analytical pipelines that turn worker conversations into structured insight about knowledge, attitudes, and behaviors—without compromising trust.
The ideal candidate combines strong engineering fundamentals with real fluency in AI-assisted development: someone who uses Claude Code and similar tools as a daily force multiplier, who has shipped LLM features to production, and who cares about building technology with and for low-income workers, immigrants, and movement organizations.
NDWA is working on the frontlines of AI development for good with many key nonprofit and Frontier organizations. This role is an exciting opportunity for an engineer who has frontend and backend experience, experimentation chops, and a massive interest in AI development, evaluation, and is ready to be a nonprofit champion for using AI to expand impact.
Primary Roles & Responsibilities
Ask Aya — Primary Engineering Ownership
- Serve as the primary engineer on Ask Aya, leading iterations, feature development, and integrations.
- Own the full LLM application stack: prompt design and versioning, offline evaluation suite, retrieval (RAG) over NDWA's knowledge base, tool/function calling, multi-turn conversation state, content moderation, and fallbacks.
- Ship features that measurably improve Ask Aya's helpfulness for domestic workers—e.g., drafting messages to employers, explaining legal entitlements, surfacing benefits eligibility, and connecting workers to local NDWA chapters and campaigns.
- Build and maintain a rigorous evaluation harness: regression tests on prompts, eval sets grounded in real worker needs, working with product team on red-teaming for safety, and bilingual (English/Spanish at minimum) quality checks learning implementation.
- Continuously improve latency, cost, and reliability of the Ask Aya runtime as well as guardrails and product features that drive engagement.
Membership Center Platform
- Contribute as a senior engineer on the broader Membership Center platform—the web/app experience where domestic workers sign up, access resources, and engage with NDWA.
- Build and maintain integrations between Membership Center, Ask Aya, and NDWA's organizing stack (EveryAction Email & CRM, SMS platforms: Mobile Commons, webpage, data warehouse, and Mobilize integration).
- Ensure authentication, accessibility, internationalization (especially Spanish), and mobile-first performance for users with limited bandwidth and older devices.
AI-Assisted Engineering Practice
- Use AI coding tools as a core part of your daily workflow—for scaffolding, refactoring, test generation, etc.
- Establish patterns, prompts, and guardrails so the rest of the team can safely and productively use AI tools in their own work.
- Stay current on the rapidly evolving AI/LLM landscape and translate relevant advances into pragmatic improvements in our stack and in our product infrastructure.
Collaboration, Craft & Culture
- Work closely with product managers, designers, data, organizers, and policy staff to turn worker needs into shippable features.
- Write clear technical specs, RFCs, and runbooks; review code with care; mentor more junior engineers and oversee contractors.
- Be a thoughtful voice on privacy, safety, and ethics—particularly as they affect immigrant workers and workers in vulnerable labor arrangements.
Qualifications
Required
- Senior-level engineering experience — 7+ years building and shipping production web/product software, with meaningful time owning systems end-to-end.
- Hands-on LLM application experience — has designed, built, and shipped LLM-powered features in production (not just prototypes). Comfortable with prompt engineering, RAG, tool/function calling, evaluation, and handling non-determinism in real systems.
- Deep understanding of AI fundamentals — can reason about model selection, context windows, cost/latency tradeoffs, hallucination and safety failure modes, structured output, and eval design.
- Fluent use of AI coding tools — uses Claude Code or equivalent daily; can describe concrete ways AI tools have changed how they ship, and can coach others to do the same.
- Strong full-stack fundamentals — TypeScript/JavaScript, Node.js, React/Next.js, REST/GraphQL, SQL.
- Data layer experience — Prisma (or comparable ORM), PostgreSQL, schema design, migrations.
- Cloud & infrastructure — AWS (Lambda, S3, RDS, ECS/Fargate or similar), CI/CD, observability, and basic infrastructure-as-code.
- Data pipelines & scheduled jobs — has built ETL or analysis pipelines (cron/EventBridge/Step Functions/Airflow-class tools), with proper retry, idempotency, and monitoring.
- LLM analytics & structured extraction — experience using LLMs to analyze unstructured content (e.g., conversations, documents) and produce durable structured outputs—topics, signals, flags, intent—that become a permanent analytical layer downstream systems can rely on.
- Framework-driven conversation analysis — comfortable designing LLM-graded prompts against a research framework (e.g., KAB—Knowledge, Attitudes, Behaviors) to evaluate conversations for things like whether a user received relevant information, showed shifts in confidence, or surfaced intent signals such as asking for help drafting a message to an employer.
- Privacy-preserving data architecture — can design end-to-end pipelines that run LLM analysis within our own infrastructure under a ZDR (Zero Data Retention) agreement, write structured outputs back to a Prisma/AWS setup, and delete source content on a defined retention schedule.
- Privacy & security mindset — experience handling sensitive user data, working within retention and -style constraints, and designing systems that minimize collection.
- Clear written communication — can write specs, decision docs, and runbooks that other humans actually use.
Strongly Preferred
- Experience with the Anthropic/Claude API — including streaming, tool use, prompt caching, and working under a ZDR agreement.
- Evaluation & analytics for LLM systems — has built eval harnesses, offline grading pipelines, or LLM-as-judge systems for quality and safety.
- Spanish language skills — either conversational Spanish or experience and comfortability building bilingual (EN/ES) product experiences.
- Mission experience — prior work in civic tech, campaigns, worker or immigrant advocacy, public benefits, or other movement/organizing contexts; or experience building technology for and with low-income communities.
- Integrations with organizing stacks — EveryAction, Mobilize, SMS platforms (broadcast and P2P), or similar.
- Accessibility & internationalization — shipped products that meet WCAG and work well on low-end mobile devices and slow networks.
What Success Looks Like in the Initial Term (June – December 2026)
- Ask Aya measurably improves on helpfulness, engagement, accuracy, and safety, with a durable evaluation system behind every change.
- Online and Offline evaluations are dependable for Ask Aya.
- Engineering requirements tied to distribution and reengagement and adoption hypothesis are scoped out through December.
- Sprint planning accurately reflects
