# Navina Ramesh

> Engineer at StarTree, Ex-Linkedin, Ex-Yahoo

Location: Mountain View, California, United States
Profile: https://flows.cv/navina

Experienced in building feed personalization system and distributed data systems; Passion for open-source software development
Familiar languages - Java, Scala, C, Python, Perl, Javascript, Python

## Work Experience
### Principal Software Engineer @ StarTree
Jan 2021 – Present

### Staff Software Engineer @ LinkedIn
Jan 2018 – Jan 2021 | Mountain View, CA

### Sr. Software Engineer @ LinkedIn
Jan 2015 – Jan 2018 | Mountain View, CA
Architecting Apache Samza - a generic stream processing framework - to be elastic and usable as a library by using Zookeeper for coordination
Design provides customizable coordination primitives that enables Samza to be executed across various deployment environments such as Yarn, containerized environment like ECS, Azure etc

### Software Engineer @ LinkedIn
Jan 2013 – Jan 2015
On Feed Personalization Infrastructure:
* Improved pagination model used in LinkedIn's homepage feed
* Enabled PYMK results to be ranked and rendered in the feed
* Integrated the Play-based homepage to use the new light-weight Feed API , which involved translating request/response data and also, asynchronously decorate the various entities in the response so that it renders in parity with the older API
* Investigated the use of Samza to setup the realtime data pipeline in feed infrastructure

On Samza:
* Migrated to use the new Java-based Kafka Producer
* Investigated and added Zopkio - an integration test framework to Samza
* Implemented local state re-use and host-affinity in Samza

### Graduate Student @ University of Florida
Jan 2011 – Jan 2013
Interested in Systems - Virtualization and Cloud Computing

Volunteer Programmer/Assistant at Center for Autonomic Computing, University of Florida

### Engineering Intern @ Facebook
Jan 2012 – Jan 2012 | Menlo Park, CA
Intern in Ads Team, Infrastructure Engineering
1. Developed an Ads Cluster validation system as the intern project (Python)
Probing the cluster nodes directly and aggregating their states for validation
The tests, verification mechanism and clusters were configurable using a JSON formatted config file

2. Automated Shard Distribution Verification in Ads Cluster Racks
Uniform distribution of Ads Database shards was required within every rack, as well across the racks in a cluster
The tool verifies the uniform distribution and on failure, alerts the authorized person
Future Goal: Verify and intelligently adjust/correct the distribution of shards, without human intervention

### Sr. Systems Engineer @ Yahoo
Jan 2008 – Jan 2011 | Bangalore
Remote administration and Management of Yahoo! Search Clusters, involving 80k+ servers (mostly RHEL machines)
Responsible for automating the application maintenance process in the production environment
System engineering of search component applications (crawlers/rule processing engine)
Scripting primarily in Perl

### Part-time programmer @ Indian Institute of Technology, Madras
Jan 2006 – Jan 2006
This summer project involved the conversion of existing Offline Data Processing System for M.Tech. Admissions at IITM to an Online system. The system was deployed for Graduate Aptitude Test in Engineering (GATE) based admissions to M.Tech. programmes in all IITs from March, 2007 and is still in use
Developed using -  HTML, CSS, JavaScript and PHP, MySQL server, Adobe Photoshop & Image Ready.


## Education
### MS in Computer Engineering
University of Florida

### B.Tech. in Information Technology
College of Engineering, Guindy, Anna University

### Kendriya Vidyalaya IITM


## Contact & Social
- LinkedIn: https://linkedin.com/in/rnavina

---
Source: https://flows.cv/navina
JSON Resume: https://flows.cv/navina/resume.json
Last updated: 2026-04-12