Senior Data Architect & Analytics Engineer

About AHL - Saaf AI

We are building the future of mortgage lending by combining cutting-edge AI with proven lending operations. Saaf AI is an fintech startup now part of American Heritage Lending, a top-10 private lender processing billions in loan volume across Non-QM, DSCR, and conventional programs, and backed by some of the largest asset managers and funds.

We are an AI-first team. Every engineer, every product decision, every workflow is designed around the question: “How does AI make this faster, smarter, or more reliable?” If you’re looking to push the limits of your expertise — using the latest AI tools and processes daily, not as an experiment but as your primary way of working — this role will put you at the bleeding edge of what data architecture and analytics engineering looks like in 2026 and beyond.

The Role

We’re hiring a Senior Data Architect & Analytics Engineer to own and evolve our analytics data platform. This is the first dedicated data hire — a hands-on builder who will shape how data powers every decision across underwriting, sales, and operations.

The foundation is in place. But “foundation” is exactly the right word — the exciting, high-impact work is ahead. You’ll take ownership of a platform with enormous room to grow: building entity resolution across fragmented real estate data, designing enrichment pipelines that turn raw data into actionable intelligence, creating the semantic layer that makes AI-powered analytics possible, and evolving the architecture as we scale into new loan programs and data sources.

This role is right for you if:

• You’ve built and owned a modern analytics stack end-to-end — not just contributed to one

• Data integrity isn’t something you think about after the fact — it’s the first thing you design for

• You’re energized by building on a strong foundation and taking it somewhere ambitious

• You want to be the foundational data person at a fast-growing company — high ownership, high impact, clear path to leading a team

What You’ll Own

Analytics Platform — Build It, Run It, Evolve It

• Own the production data pipeline end-to-end — ingesting external real estate and borrower data, transforming it through a layered model (staging → enrichment → business-ready marts), and keeping it reliable

• Build and maintain the semantic layer — model descriptions, metric definitions, and metadata that power BI dashboards, AI assistants, and self-serve analytics

• Manage warehouse infrastructure — roles, permissions, cost optimization, performance tuning

• Design and implement CDC replication to unify internal loan data with external enrichment sources, creating a single source of truth for borrower and property intelligence

Entity Resolution — The Hardest Problem Here

This is the most technically challenging and impactful part of the role. Real estate data is fragmented. You’ll design and build the matching engine that connects these dots:

• Build a production entity resolution pipeline — from deterministic exact matches to probabilistic fuzzy matching

• Calibrate match thresholds against real data and continuously improve recall and precision

• Design the feedback loop where operations teams validate matches, and those validations improve the model over time

Data Modeling & Growth

• Evolve and extend the existing models — borrower experience scoring, portfolio analysis, entity relationship mapping, lead enrichment, and scoring

• Design new models as the business grows — underwriting packages, market intelligence, next-best-action recommendations, transaction timelines

• Extend the data architecture as new loan programs, data vendors, and internal systems come online — building for configuration-driven extensibility, not one-off code changes

Data Quality & Governance

• Treat data quality as a first-class product — automated testing, validation frameworks, monitoring dashboards, and alerting

• Own the governance playbook — naming conventions, schema versioning, lineage tracking, migration processes

• Ensure regulatory compliance in data handling (FCRA, GLBA) — borrower and property data in lending carries real legal obligations

What We’re Looking For

Must-Have

• dbt + Snowflake depth: 3+ years hands-on with dbt (models, tests, macros, documentation, Cloud environments) and Snowflake (data sharing, warehouses, roles, cost management). This is the core of the job.

• Advanced SQL: Query optimization, window functions, CTEs, incremental models — you think in SQL.

• Data quality obsession: You’ve built automated testing, validation, and monitoring into data platforms — not as an afterthought but as a design principle.

• Entity resolution or record linkage: Experience with probabilistic matching frameworks, or strong willingness to go deep quickly.

• Python for data engineering: Comfortable writing data processing scripts, pipeline tooling, and working with matching/ML libraries.

• Hands-on ownership: You write SQL, build pipelines, debug data issues, and own systems end-to-end. Not a diagram-only architect.

• Startup pace: Comfortable with ambiguity, able to prioritize pragmatically, and energized by building something from the ground up.

Strong Preferences

• Experience in fintech, real estate (property records, transaction histories, valuation data), lending, or financial services

• Experience with CDC / data replication tools (Airbyte, Fivetran, or similar)

• Familiarity with BI tools (Hex, Looker, Mode) and semantic layer concepts

• Container orchestration experience (ECS/Fargate or similar) for data pipelines

• Exposure to mortgage data or regulatory compliance (FCRA, GLBA, SOC2)

• Experience designing data models that support AI-driven processes — LLM-ready data structures, feature stores, rules engines

Why choose us?

• AI-first, not AI-curious. We don’t use AI as a buzzword — it’s how we work. Engineers pair with AI daily. Product decisions are informed by AI analysis. If you’ve been waiting for a team that actually operates this way, this is it.

• Foundation laid, future wide open. The platform is just getting started — You’re not inheriting a finished system; you’re inheriting a launchpad.

• High ownership, real impact. as the first dedicated data architect hire.

• Mission that matters. Transform a $2 trillion industry and make homeownership more accessible. Your work directly impacts thousands of borrowers’ paths to homeownership.