Harshita Maddi

Software Engineer

United States6 yrs 10 mos experience

AI EnabledHighly Stable

Key Highlights

Led backend engineering for a real-time bidding platform.
Developed a hybrid ML model predicting COVID in expectant mothers.
Boosted ad inventory eCPM by 25% through innovative solutions.

Stackforce AI infers this person is a Backend-heavy Fullstack Engineer in the Ad Tech industry.

Contact

Skills

Core Skills

React.jsAlgorithmsPythonMachine LearningData Visualization

Other Skills

SQLJenkinsNeo4jApache JenaOutlier DetectionProblem SolvingComputer ScienceSumlogicCourse ManagementProject ManagementGradingJavaScriptContinuous Delivery (CD)Software DevelopmentGit

About

I am a MS Data Science student at the University of Washington (graduating March 2024), with a focus on AI and Software Systems. I have over five years of experience in the ad tech industry, where I built scalable platforms for programmatic advertising using machine learning and distributed systems. My passion is to leverage data and algorithms to solve complex business challenges and create value for customers and stakeholders. At Samsung Research, I led the backend engineering for a digital out-of-home mediation platform, enabling real-time bidding on DOOH ad inventory. I also developed a dynamic bid floor module that boosted ad inventory eCPM by 25% and ad space by 18% by counteracting bid shading. I filed a patent in India for a hybrid ML model to predict COVID in expectant mothers using smartwatch health data, and received the Samsung Young Achiever Award for my work. I am eager to explore new opportunities in the ad tech space, as it evolves rapidly with new formats, channels, and marketplaces. I am interested in product-focused roles that optimize cross-channel measurement and attribution models, or new ventures that shape the nascent programmatic spaces. With the right team, I am confident that I can deliver high-impact projects using my full-stack engineering and data science skillset.

Experience

6 yrs 10 mos

Total Experience

2 yrs 4 mos

Average Tenure

2 yrs 1 mo

Current Experience

Egencia

Software Development Intern

Jun 2023 – Sep 2023 · 3 mos · Bellevue, Washington, United States · On-site

SQLReact.js

Samsung india

3 roles

Senior Software Engineer

Mar 2020 – Aug 2022 · 2 yrs 5 mos

DOOH (Digital Out-of-Home Advertising) | Samsung SDS
Samsung Advertisement Platform

AlgorithmsSQL

Software Engineer

Jun 2018 – Mar 2020 · 1 yr 9 mos

Dynamic Bid Floor module, Ads Platform
Individually developed a business-critical ML application to counter bid shading done by DSPs during programmatic bidding. Trained a deep learning model (Tensorflow) which increased the eCPM of inventories by 25% while maintaining steady revenue as well as increasing Ad space by 18%. Backend: Python, Database: MYSQL, CI/CD: Jenkins
Developed a dashboard using Angular to enable business stakeholders around the world to manage bid floors.
Ads Supply Side platform
Worked in a team of 2 to develop short text language detection on ad titles using Naive Bayes with character ngram (accuracy greater than 94%) and categorization of creatives into IAB categories using spacy for POS tagging and Word2Vec mode
Knowledge Graph for Services
Created Demo using Neo4j for knowledge graph, Apache Jena for handling RDFs and creating Oncology using Protege. This was done to visualise and utilise benefits of graph DB for churn prediction and visualising user journey in Samsung Mall
Completed POC of a Fraud Detection in ads based on Device and Network using outlier detection techniques and linear regression models
POC on NSFW Ad Classification using one-shot learning and analysing libraries like YahooNSFW, Amazon Rekognition

AlgorithmsSQLPythonJenkins