• Reduced MultiSearch latency ~40% by redesigning OpenSearch query structure, removing redundant aggregations, and delivering a production-ready MultiSearch API.

• Architected an event-driven orchestration layer (EventBridge → Step Functions → Lambda) for indexing and data routing with strong idempotency and fault tolerance.

• Rebuilt observability for search/relevance systems, adding metrics, structured logs, dashboards, and alerts that significantly cut debugging time.

• Standardized Statsig experimentation for ranking features, adding guardrails, improving metric integrity, and accelerating iteration for DS + Product.

• Improved search relevance quality by tightening signal validation, strengthening multi-source request handling, and fixing indexing inconsistencies.

• Reduced operational load by eliminating noisy endpoint traffic, improving feature-flag rollouts, and stabilizing internal search endpoints.

• Maintained large-scale ETL pipelines (Snowflake, Airflow, Glue, Lambda, S3) powering event ingestion, feature generation, and ranking workflows.

• Partnered with Search, DS, and Product to shape ranking improvements, refine demos, write specs, and drive cross-pillar technical decisions.

• C3.ai provides scalable big data & IoT machine learning applications to enterprise customers.

• Core work involves scalable data model design (Postgres, Cassandra), analysis, cleaning (Pandas), integration (internal JavaScript frameworks) & feature development (Jupyter).

• Leverage tools like Splunk to diagnose potential environmental issues, scaling requirements, and job optimization (Batch Jobs, MapReduce etc) on complex distributed systems deployed across various cloud providers (AWS, GCP, Azure).

• SkyGrid is a Boeing, SparkCognition company that is enabling the future of urban aerial mobility.

• Implemented scalable communication logic between the client and a myriad of backend Docker based micro-services. Some of these services were traditional REST APIs while others were web sockets (Pulsar) used to track drone location in realtime etc.

• Engineered core stakeholder features such as airspace awareness, radar detection, EOS Blockchain smart contract calls, route generation, drone marker tracking on a live map (2D/3D), generating multiple map layers (weather, traffic, COVID-19 density, etc) via Mapbox API.

• Member of a small team working on IBM Cloud Datashield, software that runs on Kubernetes clusters using Intel SGX chips. Datashield enables applications to keep their data encrypted, even while it’s in use.

• Integrated the backend logic of IBM Cloud Datashield as a service within IBM Cloud using Node JS, Docker, & Kubernetes.

• Automated product delivery & integration mechanisms using Groovyscript inside Jenkins pipelines & cloud functions.

• Wrote robust unit and end-to-end testing systems in Chai & Axios.

• SnapIoT is a life sciences company specializing in developing modern streamlined solutions for the healthcare industry.

• Worked on updating several of their iOS/Android apps (Swallow-Rehapp on App Store/Google Play) as well as adding additional classes to their SDK for the snapp.click project, which is a drag and drop interface for people to make their own native iOS, Android, & Web apps without having to write any code.