Naman K.

Senior Data Engineer | SaaS | Marketing | E-commerce | Agency Infrastructure I build data infrastructure that actually gets used. Not dashboards that look impressive in a demo and collect dust after handover, but pipelines, models and reporting layers that teams rely on every single day to make real decisions. Most of my work sits across three areas. Marketing agencies where clean, accurate numbers across 40+ client accounts aren't optional because one broken connector affects dozens of accounts simultaneously and data trust is everything. E-commerce brands where true ROAS, LTV and margin need to come from one reliable place rather than five different platform reports that all disagree. And SaaS businesses where product usage data, subscription metrics and CRM pipeline need to sit in the same warehouse and tell a coherent story. What I actually build: 1. Data pipelines that run without babysitting. Airbyte, Fivetran, Supermetrics and custom Python connectors into BigQuery or Snowflake, with monitoring and alerting so I know before you do when something breaks. 2. dbt transformation layers that encode real business logic. Staging, core and mart models with tests at every layer so the numbers your team reports from are trustworthy, not approximately correct. 3. Dashboards teams actually open. Looker Studio, Metabase and Power BI built for the person making the decision, not the person who built the pipeline. 4. AI powered querying via Claude MCP. I've deployed BigQuery MCP servers so non-technical teams can ask plain English questions against clean warehouse data without writing a single line of SQL. Industries I know well: performance marketing agencies, DTC e-commerce, SaaS, telehealth and non-profits. Stack: BigQuery, Snowflake, Redshift, dbt, Airbyte, Fivetran, Supermetrics, Python, Looker Studio, Metabase, Power BI, n8n, Claude API, Next.js. If you need someone who treats data accuracy as non-negotiable and can work independently without hand-holding, let's talk.