**Data Migration / Warehousing -**- We move data from all popular platforms into Snowflake, Redshift and Big Query in no time- If the platform is not popular but adheres to standard REST API agreements, we have stock code to build a pipeline between the source and popular destinations- We enrich data in a data warehouse and send it back to popular sources using reverse ETL mechanismTools Used - To get this done in the past we have used the following tools- Hightouch for reverse ETL- Python REST API code that is approved by HevoData- SQL that can handle structured and unstructured rows of data **Data Analytics / Visualization**- We use heavy SQL and moderate python to build dashboards on almost all popular Business Intelligence tools- e.g., Looker, Power BI, Tableau, Mode analytics, Matamo etc- We have stock SQL code that can be plugged into Saas companies to derive insights- From Python standpoint we use seaborn to visualize data on cloud notebooks- We also help with embedded visualization via iframes **Product Analytics**- In most cases we use Amplitude / Mixpanel for product analytics- We provide funnel analysis- Drop off cohort analysis- User Journey over time / frequency- Pathfinder- Behavioral graph etc.

* Built live data pipelines on AWS / GCP using Python, SQL, EC2, Fivetran, DBT and Snowflake etc.* Led a team of analysts & engineers and worked with senior vice presidents to foster a culture of collaboration, innovation, and excellence.* Oversaw data collection, analysis, and reporting processes, ensuring data accuracy, quality, and security.* Collaborated with department heads to identify key performance metrics and provided actionable insights to support decision-making.* Managed data infrastructure, including data warehouses, ETL processes, and data governance, ensuring compliance with industry standards and regulations.* Presented findings and recommendations to senior management, clients, and stakeholders in a clear and comprehensible manner.* Consolidated Snowflake data warehouses and reduced the number of compute instances by 40%.* Migrated more than 1 million tasks from One Asana workspace to another using Asana API, Python Boto3, S3 bucket, and DataBricks notebook.* Ingested data from multiple sources including frontend, backend, click-stream, and 3rd party data.* Ensured ingestion process was fault-tolerant and was able to handle sudden spikes.* Cleaned, enriched, and transformed raw data as needed. Filtered out irrelevant information, and aggregated data after joining multiple streams* Implemented alerting mechanisms to notify stakeholders when specific conditions or anomalies were detected.* Set up monitoring to ensure the pipeline’s health and performance* Ensured data security and compliance with relevant regulations (e.g., GDPR, HIPPA). Implemented encryption, access controls, and auditing when necessary.* Documented the pipeline's architecture, components, and processes on Lucid chart to facilitate collaboration among team members and for future maintenance.

Build the following data stacks infrastructure* https://github.com/cdevairakkam7/Tech-Stack* https://github.com/cdevairakkam7/notes/blob/main/laptop_mobile_journey_measurement.MD* https://github.com/cdevairakkam7/notes/blob/main/attribution_model.MD* https://github.com/cdevairakkam7/notes/blob/main/cost_per_acq.MD* https://github.com/cdevairakkam7/notes/blob/main/offline_campaign_performance.MD

I built this stack during my time at Good Money -> https://github.com/cdevairakkam7/Tech-Stack to help the company scale to handle millions of users.Wrote the very first version of product events and built the event taxonomy.Procured and deployed the following Customer Data Infrastructure - SegmentProduct Analytics - Amplitude Install Attribution - AppsFlyer Split testing - Optimiezly CRM - Braze [Also recruited the CRM manager]Data Warehouse - SnowFlake BI Tool - Mode Analytics Digital Experience Analytics - Fullstory Aggregated user attributes using DBT [wiki.army/Data-Build-Tool].

I built the very first multi touch attribution model to measure the success of online and offline campaigns [Ad budget was in millions of dollars per quarter]. Online Ad Campaign Levels by Complexity → https://lnkd.dev/multi_touchOffline Marketing → https://lnkd.dev/offline_marketing. Built the very first churn prediction model using survival analysis to prevent users from churning.Migrated from Mix-panel to Amplitude. Wrote Python code to reward users with actual dollars based on conditions.Sample Data Analytics dashboards I built -Sample dashboard → https://wiki.army/sample [user name - user_ ; password-Admin12#]Infrastructure behind the dashboard → https://wiki.army/infra-behind-dashSQL → https://github.com/cdevairakkam7/SQL-Queries

I analyzed Google Indoor Maps' data using Big Query [Google's own query engine] and python. SQL → https://github.com/cdevairakkam7/SQL-QueriesI trained new recruits on how to build dashboards by combining geospatial data and financial data. Python dashboards were built using seaborn -> https://wiki.army/seabornData Manipulation was done using pandas -> https://wiki.army/pandasData Cleaning procedure -> https://wiki.army/data-cleaningFeature selection procedure -> https://wiki.army/feature-selectionI built machine learning models using the following techniques End to End process flow → https://lnkd.dev/end-to-endDecision Tree → https://lnkd.dev/decision-treeXGBoost → https://lnkd.dev/xg-boostKNN→ https://lnkd.dev/knn