Aditya Sheth

Software Engineer

Bengaluru, Karnataka, India3 yrs 2 mos experience

Most Likely To SwitchAI ML Practitioner

Key Highlights

Engineered a real-time analytics API for product teams.
Achieved 70% cost reduction in cloud infrastructure.
Standardized CI pipelines improving deployment stability.

Stackforce AI infers this person is a SaaS-focused Software Engineer with strong backend and data engineering skills.

Contact

Skills

Core Skills

Amazon Web Services (aws)Apache AirflowGo (programming Language)BigqueryAiAwsJenkinsKubernetesJavaScalaData AnalysisElasticsearchPython (programming Language)Algorithms

Other Skills

Claude AIChromaDBNAT GatewayMongoDBCassandraSparkSQLPythonRedisData StructuresNode.jsDevOpsTerraformC++Competitive programming

Experience

3 yrs 2 mos

Total Experience

1 yr

Average Tenure

1 yr 10 mos

Current Experience

Wayground (formerly quizizz)

Software Engineer

Jul 2024 – Present · 1 yr 10 mos · Bengaluru, Karnataka, India · On-site

Engineered a Go-based analytics API serving real-time funnel analysis and user flow visualisation on BigQuery, enabling product teams to analyse conversion patterns across millions of daily events with advanced segmentation capabilities, including user cohorts, A/B test splits, per-step behavioural filtering, and item-specific tracking for data-driven product decisions.
Architected AI analytics engine integrating Claude AI with BigQuery via Model Context Protocol (MCP), experimenting with ChromaDB vector search to index table schemas with intent-based embeddings for automated data discovery—achieving promising results but facing accuracy limitations in complex scenarios. Deployed production MCP server with 7 specialised tools and semantic query matching across enterprise datasets.
Led infrastructure changes for domain migration from quizizz.com to wayground.com using CloudFront and DNS, enabling seamless user redirection with under 2 minutes of downtime.
Led the migration from AWS-managed NAT Gateway to a self-hosted, highly available NAT instance, reducing total NAT Gateway expenses by 70%—from $25,000 to under $8,000 per month. This $17,000 monthly saving accounted for approximately 7% of total infrastructure costs, significantly optimizing cloud spend.
Own the Airflow infrastructure and DAGs responsible for syncing data from primary sources (MongoDB, Cassandra) to BigQuery; developed a reconciliation DAG to validate data consistency between MongoDB change streams, Cassandra CDC, and BigQuery, ensuring reliable and accurate analytics.
Standardized CI pipelines across the company, enabling a smooth Kubernetes migration and replacing unstable legacy code with a streamlined Jenkins-based process that improved deployment stability and developer efficiency.

Amazon Web Services (AWS)Apache Airflow

Media.net

3 roles

Software Engineer

Jul 2023 – Jul 2024 · 1 yr · Remote

Performed an in-depth analysis of the legacy data pipeline codebase, utilized across multiple teams, written in Java and Spark. Meticulously documented its intricacies, providing an invaluable resource for the team to streamline future development and troubleshooting endeavors.
Created an internal failure reporting tool that notably accelerated the identification and resolution of data pipeline issues, thereby improving operational efficiency across the board.
Contributed in the improvement of a custom project enabling LTV (Lifetime Value) calculation for Firefox extensions by leveraging historical data from Chrome extensions. The generated LTV insights are instrumental for business teams to conduct cohort analysis, tracking user performance over time since installation.

JavaScala

Software Engineer Intern

Jan 2023 – Jun 2023 · 5 mos · Remote

Implemented and deployed a fixed-variance variant of the Thompson Sampling algorithm to maximize advertisement revenue for Media.net’s job finding site targeting the U.S. blue-collar job market. This data-science technique enabled the team to programmatically display more popular jobs to users, thereby maximizing earnings measured by the Revenue per Impression (RPI) metric.
Contributed to developing a Job-Search Engine service, leveraging Elasticsearch for filtering jobs based on user-specified criteria like search-keywords, geo-location, and pay range. Implemented diverse strategies and a robust fallback mechanism for accurate and reliable results.

JavaSQL

Software Engineer Intern

May 2022 – Jul 2022 · 2 mos · Remote

Enhanced accuracy and efficiency of the team’s marquee product: job-search site for the US-blue collar job-market.
Improved job-search engine by devising an algorithm to extract user search term intent into multiple fields like company- name, job-title, job-type etc., and achieving 97% accuracy, while also reducing request processing time by 5x using Redis caching.
Leveraged sophisticated techniques including Bk-tree data structure and Dynamic-programming heuristic to handle spelling and spacing mistakes in user search terms.

Python (Programming Language)Algorithms

Fi

Software Engineer Intern

May 2021 – Jul 2021 · 2 mos · Remote

Created and deployed an internal failure reporting tool to validate the company database. Automated alerts on a dedicated Slack channel upon detection of issues, leading to faster error identification and reducing developer testing time during database logic changes or code interactions.

ScalaPython (Programming Language)