Mekhma Tamang

Veel Inc. Dec 2024 - Present

Specialized in converting fragmented manual workflows into high-performance, automated data systems. Bridged the gap between data engineering and campaign execution by aligning brand goals with creator relevance, engagement patterns, and audience context.

  • Problem Solving: Identifying the major data requirements of different teams, solving these requirements by validating the different data sources, making custom scrapers(eg: meta, linkedin, CRM) and integrating them into Medallion architecture. Transforming extracted data efficiently and delivering them to respective team with proper required structured format.

  • Data Integrity: Before delivery, I proposed a system that validates and cleans large datasets against predefined company standards to ensure high reliability for downstream analysis and campaign planning.

  • Cross-functional Collaboration: Worked on multiple domain of data needs. For eg: I assisted AI/ML team by providing the production-level light-weight custom scraper, that scrapes the dynamic semi-structured data and transform them into ready-to-feed structured data. This enhances Veel’s features by 50x faster and maintaing 40% more accuracy.

  • Executive support: Cleaned and Validated data resides in gold layer of our system but lack of sales team SQL knowledge, they weren’t able to use that efficiently. To tackle this, I proposed a RAG system based on vanna framework which allows sales person to chat with data. This system empowered sales team with data transparency to seal the deal. The problem’s solution was totally research and proposed by me.

  • Incremental Quality Over Quantity: Handling over a millions of raw data across multiple sources. Batch processing to analyze and extract qualified data for delivery.