Sahil Shah

Senior Software Engineer

Mumbai, Maharashtra, India6 yrs 2 mos experience

Most Likely To SwitchHighly Stable

Key Highlights

Proven track record in improving system reliability.
Expertise in real-time bidding and cloud solutions.
Strong foundation in machine learning and data integrity.

Stackforce AI infers this person is a Backend-focused Software Engineer with expertise in Cloud Computing and Ad Tech.

Contact

Skills

Core Skills

MysqlData IntegrityDatastreamGcpReal Time BiddingDistributed SystemsFraud DetectionMachine Learning

Other Skills

AlgorithmsApache SparkAvailability MetricsC (Programming Language)C++Cloud ComputingCloud DevelopmentComputer ArchitectureData StructuresDesign DocumentationDockerDocker ProductsFeature EngineeringFlaskGTID Source

About

Builder. I love Zero to One work. I spend my free time learning about the startup ecosystem in India and beyond, my favourite podcasts include: Acquired - Every Company Has A Story Masters of Scale with Reid Hoffman The BarberShop with Shantanu If you're interested in building a tech startup I'd love to chat. Feel free to hit me up.

Experience

6 yrs 2 mos

Total Experience

1 yr 6 mos

Average Tenure

3 yrs 11 mos

Current Experience

Google

3 roles

Senior Software Engineer

Apr 2025 – Present · 1 yr 2 mos

Software Engineer III

Promoted

Sep 2023 – Apr 2025 · 1 yr 7 mos

TL for MySQL Source:
Triaged customer issues, identified and fixed deep data integrity issues in MySQL Source. Losing customer data is intolerable, data fidelity is P0.
Improved Availability metrics from 99.8% (1st week of Sept 2023) to 99.99% (1st week of Jan 2024).
Improved JSON support in MySQL source (the biggest source of customer issues ~Sept 2023).
Testing improvements: Moved testing to incluster DB, reduced test failures due to network flakiness, drove down testing costs, and reduced testing time by 20%.
MySQL GTID Source: Enterprise customer encounter reliability problem with binary log (zero downtime failovers in managed cloud databases) -> Research and write 1 pager proposing GTID source -> design doc -> Build and ship the source in preview with 2 new grads within 6 months -> Help onboard initial customers -> GA -> Profit!

MySQLData IntegrityAvailability MetricsJSON SupportTesting ImprovementsGTID Source

Software Engineer III

Jul 2022 – Sep 2023 · 1 yr 2 mos

Building SQL Server Source for Datastream and DMS. Working in Datastream Sources India team.
Datastream: Serverless, simple CDC and Replication service. We primarily replicate data from customers' source databases onto Google BigQuery enabling customers to leverage the GCP Data Cloud.
DMS: Simple minimal downtime migrations onto GCP Databases. We're building for homogenous migration from customers' Sql Server instances to GCP's managed Cloud SQL offering for SQL Server.
In order to deliver a high fidelity experience, I'm in parallel building an integration testing framework for Google3 built services running on GCP/GKE.

SQL ServerDatastreamGCPIntegration Testing Framework

Media.net

Software Development Engineer

Jul 2021 – May 2022 · 10 mos · Mumbai, Maharashtra, India

Worked in the Online advertising (Real Time Bidding) team in a highly scalable, low latency environment bidding in 25Bn+ ad auctions per day on a multi-region distributed cloud architecture.
Designed and developed a distributed real time solution to track bidding revenue and expenditure on several business aggregate levels, saving thousands of dollars per month.
Improved performance on sections of semi-exploration traffic, doubled revenue rate, improved profit rate by 50%.
Built a machine learning model, with a low latency pipeline for click through rate prediction on premium customer traffic.
Contributed to design (design patterns, reliability) of existing and new systems.
Go | Java | Python | Kafka | ML

Real Time BiddingDistributed SystemsMachine LearningJavaPython

Paypal

Software Engineering Intern

Feb 2021 – Jun 2021 · 4 mos · Hyderabad, Telangana, India

Working in the Data Engineering Team.
Optimised Rule Recommendation Algorithm for Fraud Detection both in terms of speed and precision.
Spark POC - Implemented rule recommendation algorithm on Spark cluster using Spark's ML Lib.
Refactoring, testing, dockerizing, and improving documentation on Rule Recommender component for better developer experience for the team.
Designed a service to measure the effectiveness of Rule Recommender. Quantified resources used, benefits to merchants, and efficacy of the Rule Recommendation algorithm.
Python | Spark | ML | Docker | Java | REST
Secured a full-time offer.

Fraud DetectionSparkDockerJavaREST

Media.net

Software Engineering Intern

May 2020 – Jun 2020 · 1 mo · Mumbai, Maharashtra, India

Worked in the Dynamic Margin Management Team.
Feature Engineering for ML models used in Dynamic Margin Management(DMM) for Real-Time Bidding.
Evaluated feature encoding schemes to reduce model size and training time, without sacrificing much accuracy.
Built a simulator and dashboard in Flask to train, evaluate and compare models built on a user-chosen subset of all available features.
Python | Flask | ML | CatBoost | JS
Secured a PPO

Feature EngineeringMachine LearningFlask