Jun 2023 – May 2025 · 1 yr 11 mos · United States · Hybrid

• Engineered AI document generator for PMs, used by 1000+ users. Ingested data from GSuite, Jira, Slack and others using advanced RAG and LLMs to create reliable PRDs and one-pagers.
• Orchestrated full deployment pipelines; implemented advanced monitoring and automated feedback loop for continuous LLM quality improvement, significantly boosting performance and reliability.
• Implemented AI voice assistant for restaurant order-taking; achieved 95% call accuracy and sub-second latency across 3k+ calls using Mixture of Experts (MoE) architecture, semantic caching, and fine-tuned Llama/Mistral LLMs.
• Developed AI technical interviewer with MoE architecture, seamlessly conducting 60-minute interviews, streamlining candidate preparation and company’s hiring process.

Jun 2022 – Jun 2023 · 1 yr

• Spearheaded the "Wishmas" campaign, leading a team of 8 engineers to drive $xM in daily GMV and achieve 100 million MAU, earning top accolades from the CEO for outstanding performance.
• Orchestrated complex projects across 6+ teams, delivering key initiatives in just three months, architecting system to ensure low latency and high reliability.
• Managed top-of-funnel initiatives for Wish across all platforms, including the homepage used by 3M daily users, boosting engagement and enhancing platform effectiveness.
• Formulated long-term vision and KPIs for team projects, spearheaded talent acquisition (grew team from 2 to 8 engineers), and mentored the team to drive peak performance leading to promotion of multiple engineers in the team.

Oct 2019 – Jun 2022 · 2 yrs 8 mos

• Rearchitected the Wish homepage as android tech lead, reducing latency by 20%, accelerating feature development (3 months of work in 2 weeks); resulting in GMV increase by 4%.
• Engineered a live, video-based shopping MVP with integrated live chat on Android, persuading leadership to invest further by presenting a compelling business case.
• Implemented a microservice for video uploads on Kubernetes, boosting p99 performance by 30% and cutting error rates by 15% through advanced tracking, monitoring, and alerting with Prometheus, Grafana, and PagerDuty.
• Optimized database queries to improve latency by up to 10x and established robust standards to prevent future performance issues.
• Architected user-generated content system, leading a team of 4 engineers to enhance user trust and product discovery, resulting in a 12.5% increase in customer engagement through reviews, photos, and videos features.

May 2019 – Aug 2019 · 3 mos

• Developing an e-commerce system to support 10,000 requests per second with an uptime of 99.99% and an ability to handle 2.25 million products.
• Designing a distributed system to support high levels of performance, availability, scalability and security.
• Deploying an Elastic Load Balancer with a Kubernetes cluster to create a group of fault tolerant servers with an ability to scale automatically to handle dynamic load.
• Technologies: NodeJS, Express.Js, MongoDB, REST API, Artillery, Redis, Jenkins, Docker, Kubernetes, AWS

Jan 2019 – Aug 2019 · 7 mos

• Developing a radioactivity analysis tool for Department of Energy in collaboration with CMU’s Robotics Institute to safely and quickly analyze uranium contamination in the pipes present in nuclear facilities across the nation.
• Extending and implementing the requirements of the current system to facilitate its deployment in production.
• Reducing 75% of time required in measuring and analyzing contamination in pipes using an automated system.
• Technologies: Apache Tomcat, Java, Jersey, JAX-RS, PostgreSQL, Gradle, JUnit, Mockito, TestNG, Jenkins, SonarQube, Vagrant, Automated Testing, Continuous Integration and Deployment, Agile Methodology.

Dec 2018 – Apr 2019 · 4 mos · Greater Seattle Area

• Designed and developed an architecture solution for Anti-Bullying Report and Support system covering requirements, context, component, security, data model, and deployment architecture.
• Addressed architectural solution facilitating anonymity, privacy, security, low running cost among others.
• Secured 2nd position in a national level architecture competition among 51 teams.

Aug 2017 – May 2018 · 9 mos · India

• Led the development of Backend as a Service for automated REST API server code generation and Analytics as a Service to generate Python Flask server code for data analysis.
• Automated development of user application on NodeJS, Python Flask, and MongoDB without coding.
• Implemented Flask code generation for analytics and designed architecture for storing and mapping user application details to generate server code.
• Technologies: NodeJS, Python Flask, MongoDB, and scikit-learn

Jun 2017 – Sep 2019 · 2 yrs 3 mos · India

• Cofounded full stack Android application to help non-native English speakers learn English words which has over 45,000 worldwide users.
• Developed MVP architecture for Android application reducing code by ~30% and app's time to market by ~25%.
• Designed communication interface between frontend and backend, created backend server on NodeJS, developed quiz feature, and managed infrastructure on AWS.
• Won national level start-up competition among 400+ teams in student category.
• Technologies: Android, NodeJS, MongoDB, Firebase Analytics, Amazon AWS, Graph API, Dagger, Retrofit, RxJava.

Apr 2017 – Jun 2017 · 2 mos · India

• Developed an Android application for millions of followers of Brahmakumaris. Features such as live videos, audios, articles, gallery, and daily thoughts were present and were curated by Brahmakumaris.
• Developed a complete backend server and web-based admin panel for representatives.
• Technologies: Android, NodeJS, MongoDB, HTML, CSS, jQuery, Ajax, Retrofit.

Jun 2016 – Jul 2016 · 1 mo · India

• Developed Bill of Materials (BOM) auditing software to audit spreadsheets of materials used in the creation of various mechanical parts.
• Reduced auditing time from 1.5 hours to 15 seconds and solely created this system.
• Technologies: Java, JavaFX, Apache POI library