Signal 01
40 %
Manual tasks automated
1 h saved every morning
At Hoist Finance, automating 40% of manual tasks saved several dozen colleagues an hour every morning.
Available from October 2026
Mickaël Rakotoarinivo
Data Engineer in Lausanne.
5+ years designing, deploying and operating data platforms for finance, logistics and analytics teams.
- Azure Databricks
- Snowflake
- Python
- PySpark
- SQL
Daillens, VD
- Azure Databricks
- Snowflake
- Python
- PySpark
- SQL
Make data usable. Keep it reliable.
My work starts with imperfect sources and ends with data that business teams can use. Along the way, I model, test, monitor and document it.
I work on batch and streaming pipelines, medallion architectures, and star and snowflake schemas. I also handle day-to-day production operations: incidents, performance, on-call support and knowledge sharing.
Measured impact
Time saved, every day.
Signal 02
+20
BI reports delivered
At Mondial Relay / InPost, over 20 reports built for operations, customer relations and Damage & Loss, used by around 30 people.
Signal 03
2 h→20 min
Databricks workflows optimised
Around 83% less compute time on selected workflows. At Mondial Relay / InPost, data modelling and Spark SQL / PySpark optimisation cut their runtime from about 2 hours to 20 minutes.
Production experience across three businesses.
07.2026 → Present
API SA, assignment at tl
Data Engineer / Analytics Engineer
Modernising finance and BI data on Snowflake. SQL and Python ELT pipelines, dimensional models, tests and a dbt-inspired execution framework. Around ten raw stored procedures transformed daily into Silver tables for tl's finance leadership and management accounting teams.
09.2022 → 06.2026
Mondial Relay by InPost Group
Data Engineer
Batch and streaming pipelines on Azure Databricks and Delta Lake. Unity Catalog, CDF, Auto CDC, Kafka, SCD, snapshots and GitLab CI/CD in an international team.
05.2021 → 08.2022
Hoist Finance
Data Analyst
Automated data extraction and quality checks, scheduled ETL pipelines, operational reporting and statistical analysis in banking.
From pipeline to product.
PriceHunter
A basket optimisation app that collects supplier prices, normalises products and allocates line items according to available offers.
- My contribution
- Displaying prices, stock levels and sales units on each basket line item.
- Technical choice
- A Playwright worker for supplier interactions, a CSV export fallback and human approval before sending.
Bitcoin Analysis Dashboard
A Python and PostgreSQL pipeline, dbt transformations and an interactive Next.js dashboard, updated daily with Bitcoin market data.
- My contribution
- Connecting daily ingestion to price and indicator tables, then displaying them in the dashboard.
- Technical choice
- Rebuilding dbt tables to carry historical corrections through the pipeline. A full refresh suited to the current data volume.
Open to opportunities in French-speaking Switzerland