g
geekyarun

Arun M

@geekyarun

Data Engineer and AI Systems Builder, 10 Years Experience

India
Engels
Sommige informatie wordt in het Engels weergegeven.
Over mij
I'm Arun, a Data Engineer & AI Consultant with 10+ years of experience building scalable data platforms, analytics solutions, and AI applications. I specialize in ETL/ELT pipelines (Airflow, dbt, Spark, Databricks), cloud data platforms (GCP, AWS, Azure, Snowflake), and AI-powered RAG chatbots. Formerly at Acko for 4 years, I now lead Vyntics, helping businesses turn complex data into reliable insights and intelligent products. I focus on delivering production-ready solutions that are scalable, maintainable, and business-driven.... Lees meer

Skills

g
geekyarun
Arun M
offline • 
Gemiddelde reactietijd: 1 uur

Bekijk mijn diensten

GPT-apps op maat
I will build a custom ai chatbot trained on your own knowledge base
VanafUS$ 40 / projectMeer informatie

Werkervaring

Patreon

Consultant - Data Engineer

Patreon • Freelance

Dec 2025 - Present • 6 mos

Salesforce + Braze to Databricks — Medallion Architecture Pipeline Built and maintained production data pipelines for a subscription platform, integrating two key customer-facing systems — Salesforce (sales/CRM data) and Braze (marketing/messaging data) — into a unified Databricks lakehouse. What I built: Ingestion pipelines pulling data from Salesforce and Braze APIs into Databricks, handling schema differences and API rate limits across both sources Medallion architecture (Bronze → Silver → Gold) to structure the lakehouse: Bronze: raw data landed as-is from both source systems Silver: cleaned, deduplicated, and standardized into a unified schema Gold: business-ready tables joining Salesforce and Braze data for downstream reporting and analysis Multiple transformation jobs to reconcile data models between the two systems — Salesforce and Braze structure customer and engagement data very differently, so this required careful mapping logic rather than a simple merge Worked independently to ramp up on Salesforce data concepts I hadn't used before, then built the engineering work around that understanding