SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Koltin is redesigning elder and disabled care in Mexico, offering health memberships that integrate preventive clinical care, major medical expense coverage, and wellness/community programs. The company serves adults aged 50–84 and is growing rapidly across the country.
You will join as an Analytics Engineer focused on data platform infrastructure. This is a hands-on, growth-oriented role where you build the "how" — modeling dbt tables that the entire organization relies on, and sustaining the data platform day-to-day (ingestion, orchestration, quality, documentation). You work with autonomy over your domain, paired with senior code review and mentorship. You sit close to product and business teams, learning to convert ambiguous questions into robust models.
Koltin's ambition is to move from answering known questions well, to anticipating questions, to running predictive models in product. This role builds the foundation for that leap.
**Key Responsibilities:**
- Model in dbt: build and maintain models following team layers, conventions, and dimensional modeling, with tests, documentation, and CI-passing PRs.
- Implement data contracts: translate agreements with domain-owning teams into code; detect when metrics shift meaning.
- Operate the data platform: manage Airbyte connectors, Dagster orchestration jobs, monitoring, and incident response so anomalies alert before users report them.
- Enable self-service: maintain and extend the semantic layer and models in Omni and Hex so users can answer their own questions.
- Build reliable deliverables: dashboards and reports for business and external partners, with appropriate detail and validation.
- Work embedded with product/business teams: understand the real processes behind the data.
- Participate in standards: code review, pairing, and improvement proposals; you don't set the bar alone, but you sustain and push it.
**Working Style:**
Two modes: self-service (Hex, Omni in Slack, Claude, generative AI) for rapid hypothesis testing; and built deliverables when confidence must be total. Every request leaves capacity behind—models and contracts that prevent rework. Data embeds with teams, not the reverse. Real ownership with minimal hierarchy and space to decide how things are built, with guidance on big decisions.
**What This Role Is NOT:**
- Not a BI/dashboards-only role; the heart is modeling and platform.
- Not a query help desk or central reporting team.
- Not a junior support role; you own and close complete tasks.
**Tech Stack:**
PostgreSQL, dbt, Dagster, Airbyte, Python, Omni, Hex, AWS, Git/GitHub Actions, Slack, Notion, Claude.
**Requirements:**
- 2+ years in data (analytics engineering, data engineering, or BI with modeling), having owned end-to-end work.
- Solid SQL and clear analytical modeling foundations: dimensional modeling, complex joins, incremental loads, idempotence, history management.
- Data transformation experience with dbt or equivalents: models, tests, documentation, repo work with PRs and CI.
- Python (or equivalent) for scripts, automation, and pipelines, with basic engineering practices: Git, PRs/code review, testing.
- Experience with an orchestrator (Dagster, Airflow) and/or managed ingestion (Airbyte, Fivetran or equivalent)—even maintaining existing systems.
- PostgreSQL and/or an analytical warehouse (Redshift, Snowflake, BigQuery).
- Built dashboards or BI models for self-service consumption (Omni, Looker, Hex, Metabase, Power BI, or others).
- Business communication: ask the right questions to understand what is being measured before modeling.
- Openness to giving and receiving technical feedback; this role grows through code review.
- Native Spanish and reading-level English (documentation and tools in English).
**Nice-to-Have:**
- GCP or AWS experience; infrastructure-as-code (Terraform, Pulumi, CDK).
- API consumption or building (REST, GraphQL).
- Clinical or health data experience with privacy and compliance requirements.
- OLAP architecture knowledge (MOLAP, ROLAP).
- Support for Data Science/ML work: training datasets, features, predictions served to product.
- Startup growth context: high ambiguity, low bureaucracy, shifting priorities.
