Leon Capital Group is a multi-billion-dollar holding company that owns and operates businesses across healthcare, real estate, and financial services. We found, acquire, and scale companies for the long term, backing them with shared capital, talent, and infrastructure while giving each the room to operate and grow. Our portfolio spans provider groups and patient-care platforms, real estate development and investment, and a growing set of financial services businesses. This role reports to the Chief Innovation Officer and supports every business in the portfolio.
About the Role
This is a senior, hands-on data engineering role where the Senior Data Engineer owns the enterprise data foundation AI is built on. Across our portfolio, the data maturity and structure vary: the data exists, but it is often fragmented, duplicated, or unreliable — and no model or decision tool is better than the data beneath it.
You will design platforms that serve multiple operating companies, regulated workloads, and production AI systems: shared canonical models, cross-company identity resolution, governed access, and the reliability, auditability, and cost discipline an enterprise platform demands. As we acquire businesses, you will also lead the data side of integration, folding new systems into the shared foundation quickly and retiring redundant cost.
You will be part of an extremely lean team that operates like an early-stage startup inside a larger enterprise: most of what you build has never existed here before, and you will take it from zero to production. There is no platform team, no data architect, and no one to hand the messy part to — you will scope the work, stand up the infrastructure, build the pipelines, and operate them. In exchange you get unusual latitude: direct access to the Chief Innovation Officer, no layers between you and the business owners, and the ability to set the patterns everyone who comes after you inherits.
AI is how a team this lean moves at this speed. We expect you to already use AI tools every day, coding assistants, agents, and LLMs, to write and review code, draft models and tests, profile data, map schemas, and document your work. We expect you to do it with judgment: you verify what AI produces, own everything you ship, and know when not to use it. Volume of AI-generated work is not the goal. Unreviewed output and blind trust in a model are not acceptable here, and critical thinking matters more to us than any tool.
We want someone who treats the data layer as the thing that makes AI possible — the engineer who knows what a model needs from the data beneath it, can sit with an AI Engineer and design the feature table before the model exists, and treats dirty, duplicated data as the most important problem in the building, because every segment, attribution number, and prediction downstream depends on getting it right.
Key Responsibilities
- Stand up and own the curated data layer that AI and ML depend on: clean, well-modeled marts that feed lead scoring, attribution, next-best-action, and other predictive workloads.
- Architect enterprise-grade data platforms that span multiple operating companies: multi-entity canonical models, cross-company identity resolution, role-based access and auditability, and the SLAs production AI and regulated reporting depend on.
- Partner directly with AI & ML engineers, business stakeholders, and the production AI agents and automations already running in the portfolio as the person who builds what they consume — feature tables, point-in-time-correct training sets, and serving paths — and hold the line on consistency between training and scoring.
- Consolidate and harden existing cloud data platforms: re-enable broken syncs, close single points of failure, and bring infrastructure up to architecture and security standards.
- Design and own the canonical data model and curated marts, built to remain ours regardless of which vendor or CRM sits on top.
- Build ingestion pipelines that pull from heterogeneous, often hostile sources into our schema with full provenance — vendor servicing output, API feeds, and flat-file or SFTP partner feeds with no clean API.
- Run cleansing and de-duplication as a continuous discipline rather than a one-time pass: standardize and validate source data on ingestion, match records across systems, and keep the resolved record clean so attribution and downstream models stay trustworthy.
- Build validation, monitoring, and lineage so data-quality issues are caught before they reach models, reports, or decisions.
- Treat the platform as a reusable pattern, standing up each new business's data layer rather than a bespoke build each time, so the foundation scales across the portfolio.
- Lead data integration for newly acquired businesses: assess the inherited stack, migrate what matters into the shared foundation, and retire redundant systems and cost post-close.
- Set the engineering standards the team builds against: modeling conventions, data contracts, CI/CD and testing for data, observability, and cost management.
Qualifications
Required:
- Startup and 0-to-1 build experience (required). 7+ years building data systems end to end, including time at a startup or in a 0-to-1 build environment where you took a data platform from nothing to production, as the only data engineer who owned the whole function. Being an early or founding data hire, or owning a data domain end to end at a small company, matters more to us than seniority on a large team where a platform org handled the infrastructure.
- Enterprise-scale platform delivery. You have also designed or led data platforms at enterprise scale: multiple business units or entities, large and heterogeneous source estates, formal governance, and production SLAs. We want startup speed paired with enterprise rigor, and you can point to both.
- AI-augmented daily practice. You already use AI to accelerate your engineering work and can show how: what you delegate, how you verify it, and where it got something wrong and you caught it. You treat AI output as a draft to be challenged, not an answer to be trusted. Critical thinking comes first; tool fluency comes second.
- Delivery experience in two or more industries. You have had to learn a new domain's vocabulary, constraints, and data quirks from scratch, and you have done it more than once. Healthcare, real estate, financial services, or insurance exposure is ideal, but breadth across any two unrelated verticals is what we are after.
- A track record enabling AI and ML. Hands-on experience laying the data groundwork for models — feature engineering, training-set construction, feature stores, or building the curated layer a model consumed — or a clear record as the data engineer an ML or data science team depended on. Either direction counts. What matters is that you understand what models need from the layer beneath them and can name specifically what you built.
- Strong data engineering fundamentals: schema design, ETL and ELT pipeline architecture, and data-quality and de-duplication discipline.
- Deep SQL and warehouse experience. Our current platforms are Azure-anchored (Azure SQL, dbt for curated marts), and you are comfortable working in and consolidating this stack; you are equally comfortable extending into a multi-cloud environment (we use a mix across Azure and AWS) as the work broadens across the portfolio.
- Proven experience integrating against messy enterprise and vendor systems, and against flat-file or SFTP feeds with no clean API.
- Comfort owning infrastructure without a platform team. You can stand up and run the warehouse and pipelines yourself in a major cloud environment, including your own CI, secrets management, and access controls.
- Comfort building under regulatory and fiduciary constraints, and handling sensitive, regulated data (PHI, HIPAA, financial) with the secure practices it requires.
- Excellent judgment under ambiguity and the ability to prioritize across multiple businesses without close oversight.
- A bias toward data quality: you treat trustworthy, well-resolved data as the precondition for everything downstream, not an afterthought.
Note to Search Firms and Agencies:
Leon Capital Group does not accept unsolicited submissions from search firms or employment agencies. Any resume submitted to any employee of Leon Capital Group without a valid written search agreement in place will be deemed the sole property of Leon Capital Group, and no fee will be paid in the event the candidate is hired.