Karan Ganju

Product Engineer

Mountain View, California, United States8 yrs 2 mos experience

Most Likely To SwitchHighly Stable

Key Highlights

Led cross-functional teams to build impactful ML models.
Developed underwriting systems for diverse capital products.
Enhanced ad bidding models, achieving significant performance improvements.

Stackforce AI infers this person is a Fintech and AdTech professional with strong expertise in Machine Learning and Software Engineering.

Contact

karanganju9@gmail.com LinkedIn

Skills

Core Skills

UnderwritingData ScienceSoftware EngineeringMachine LearningProduct Design

Other Skills

AlgorithmsBashBatch ProcessingCC++CollaborationComputer SecurityCryptographyData AnalysisData StructuresDecisioningDeep LearningJavaLinuxModel Serving

About

Driven staff software engineer passionate about building impactful solutions from the ground up. At Parafin, my team built and powers the underwriting system and application-time decisions for all capital applications. Past experience in Ads and Machine Learning, always eager to learn and grow.

Experience

Parafin

2 roles

Underwriting & Decisioning TLM

Promoted

Feb 2024 – Present · 2 yrs 1 mo · San Francisco Bay Area

Work closely with Data Science to guide Parafin's risk/origination posture through underwriting and decisioning across 50+ partners, different capital products (loans, spend cards and BNPL) and different data streams. Led team responsible for building and powering the
Underwriter, running for batch and realtime for millions of merchants across various capital products
Decisioning platform, running various critical pre-funding application checks in realtime

Data ScienceUnderwritingDecisioningRisk Management

Senior Software Engineer

Apr 2022 – Mar 2024 · 1 yr 11 mos · San Francisco Bay Area

Built both the batch and realtime underwriting platform serving all of the capital platforms and allowing close collaboration and fast velocity from the data science team

Batch ProcessingReal-time SystemsCollaborationSoftware EngineeringData Science

Google

Software Engineer

Jul 2018 – Feb 2022 · 3 yrs 7 mos · San Francisco Bay Area

Led cross-functional team (5+) across DS, Eng, and PM to build ML models for advertiser metric extrapolation. Owned product design, OKRs, and POC TensorFlow model; defined evaluation metrics and onboarded 3 Eng + 1 DS, improving AUC by 42%. Built scalable model serving infrastructure.
Drove ads bidding model upgrade with Google Brain, migrating from logistic regression to neural network. Delivered 78% AUC lift, 2–3× latency reduction, and 50% faster training while enhancing reproducibility and reducing data bias. Designed and implemented full training and serving stack.
Owned $1B+ ARR advertiser metric reporting pipeline; primary contact for customer escalations, feature requests, and internal product onboarding.

Machine LearningTensorFlowModel ServingData AnalysisProduct Design

Twitter

Machine Learning Intern

May 2017 – Aug 2017 · 3 mos · San Francisco Bay Area

Improved tweet search results by adding a predicted text relevance score for each tweet given the query text and using the score to rank/demote the tweets.
New search results contained 55% lesser bad search results as per human quality evaluation.

Search AlgorithmsText RelevanceMachine Learning

University of illinois at urbana-champaign

Digital Forensics Teaching Assistant

Aug 2016 – Dec 2016 · 4 mos · Urbana-Champaign, Illinois Area

Lab Assistant

Samsung electronics

Security Analyst Intern

May 2015 – Jul 2015 · 2 mos · Suwon, Gyeonggi-do, Korea

• Analyzed functionality of Device Guard, a software whitelisting security tool, to identify dependencies and possible limitations

Tinyowl

Backend Developer Intern

May 2014 – Jul 2014 · 2 mos · Mumbai Area, India

• Recreated and enhanced the management panel based on Ruby on Rails, PostgreSQL and MongoDB to provide more intuitive and comprehensive application usage statistics