Praveen Kumar

About

I'm a backend engineer at Atlan, a metadata platform for the modern data stack. My work is the plumbing underneath a data catalog: connectors that pull schemas, lineage and usage out of databases, streaming platforms and BI tools, and keep that extraction reliable at enterprise scale.

I like connectors that are boring in production: predictable memory, clear failure modes, and runs that behave the same on a demo database and a large enterprise warehouse.

Before industry I did research in parallel computing and image processing, with three papers at IEEE and international conferences, and spent a stretch deep in machine learning through the #100DaysOfMLCode challenge.

Experience

1. Present

   Senior Software Engineer · Atlan

   Metadata extraction for databases, streaming platforms and BI tools.

   • Build and maintain source connectors for relational databases, streaming platforms and BI tools, including lineage extraction.
   • Own connector reliability end to end: workflow debugging, memory and performance profiling, and root-cause analysis on large extractions.
   • Work on Atlan's app SDK–based connector platform, running on Temporal and Kubernetes.

   • Java
   • Kotlin
   • Python
   • SQL
   • Data lineage
   • Temporal

Projects

1. 2017

   Iris-based attendance system · Smart India Hackathon

   Android app using iris recognition to record attendance for MGNREGS rural-employment workers. Built for Smart India Hackathon 2017, billed as the world's largest hackathon.

   • Android
   • Computer vision
2. 2016–17

   Image inpainting cloud app

   Web application that removes unwanted objects from photos and fills the gap plausibly.

   • Image processing
   • Web
3. 2015

   Social network graph processing

   C++ analysis of 500,000 call records, modelling customers as a graph and ranking the top VIPs by connectivity and call duration.

   • C++
   • Graphs

Publications

1. 2021

   New Hybrid Approach for Exemplar-Based Image Inpainting

   3rd International Conference on Computational Intelligence in Pattern Recognition (CIPR)

2. 2016

   Analysis of Optimize Pairing for Large Scale Crew Scheduling Problem

   6th IEEE International Advance Computing Conference (IACC)

3. 2016

   Implementation of Cubic Spline Interpolation on Parallel Skeleton

   6th IEEE International Advance Computing Conference (IACC)

Skills

Languages

Java, Kotlin, Python, SQL, C++

Data sources

Relational databases, streaming platforms, BI tools

Platform

Temporal, Kubernetes, Argo Workflows

Education & learning

1. Alma mater

   National Institute of Science and Technology

   Member of Club Excel, the programming club.

• Neural Networks and Deep Learningdeeplearning.ai · 2018
• Improving Deep Neural Networksdeeplearning.ai · 2018
• Structuring Machine Learning Projectsdeeplearning.ai · 2019
• Intro to TensorFlow for AI, ML & DLdeeplearning.ai · 2019
• Machine LearningCoursera · 2017
• Python 3 Programming SpecializationCoursera · 2019
• Python for Applied Data ScienceCoursera · 2018
• Introduction to Oracle 11g: SQL2014