Aniket Bhatnagar

Led and built all of USDC stablecoin's back-end infrastructure which included:

1. Reviewing the USDC ERC-20 contract
2. Integrating with banks to receive and send dollars that can be converted to/from USDC tokens onchain (fiat to onchain token bridge)
3. USDC token wallet to send and receive USDC ERC-20 tokens
4. Integration with ERC-20 smart contract for sending, receiving, minting, and burning USDC tokens. Handle all typical problems encountered with chain integration - re-org, detecting duplicate ERC-20 transfer due to reorg, rebroadcasting stuck transactions while still ensuring no double-spends, etc.
5. Auditing to ensure that tokens on chain never exceed the money in the bank.

AOL platforms is in the business of programmatic advertising and works with leading brands to deliver their message to a wide range of audience. The user targeting service is an API used by ad servers to determine the campaign that needs to be shown to the user and guess bid value. The user targeting service has tight SLA (less than 10ms latency with 99.9% availability) and high throughput (> 500k TPS peak). As part of the project, I successfully delivered critical pieces like caching and serving segments for users that are linked with multiple browsers/devices.

Apart from building high throughput systems, I also worked with massive datapipeline for TBs of segmentation data using Spark and AWS Datapipeline

Tech stack: Java 8, Netty, Chef, Functional Programming, Spark, AWS, AWS CloudFormation, AWS ELB, AWS AutoScaling

Involved in building a data platform for Internet of things (IoT) applications that allow for necessary tools to sense/collect data, perform analysis, and build models.

- Architect to build a high-level view of the platform and necessary technology recommendations. The work was presented by our CTO at Strata Conference in London (http://www.slideshare.net/OReillyStrata/sapientnitro-stratapresentationupload) (slides 30 - 33)
- Implemented crucial pieces of the platform like real-time streams (using web sockets, Redis, and Play! Iterates), API for time series data persistence (using HBase), modular components using Cake patterns, remote widgets to capture data from IoT devices, etc.
- Traveled to Chicago to meet the data analyst team.

Tech stack: Scala 2.10.3, Play! Framework 2.2.0, HBase, Redis, Kafka 0.8, Storm, HDFS, Hadoop, Elasticsearch, StatsD + Graphite (for monitoring), Spark 1.x

NASCAR and SapientNitro had joined hands to develop a fan engagement platform that will be used by NASCAR as well as sold as a managed offering to other sports entertainment businesses.

- Team lead for integration and security track and junior architect for addressing scalability and performance.
- Evaluated various NoSQL databases for the choice of key-value and document-oriented databases. Finally closed down with Redis and MongoDB.
- Implemented a highly scalable (2 million hits per hour) and available (99.99%) OpenID 2.0 compatible identity provider solution using LDAP on the back-end and sharded Redis instances for session and volatile data store.
- Travelled to the US (Charlotte, NC) to help with deployment and cut-over activities.

Tech stack: Java/JEE, OSGi, Apache Camel, Talend, Adobe Day CQ (CMS), Spring, OpenID (provider implementation), Spring, Redis (Cache/Session/Association/AccessToken store)

Debugged performance issues in a data pipeline that had been migrated to AWS and Spark.

Contributed bug fixes in Spark-Kinesis integration and dependency conflicts in the Spark core. Also, contributed enhancements to AWS EMR Spark Bootstrap script.

Selected contributions: SPARK-8895, SPARK-7788,
SPARK-5164, SPARK-5143, HBASE-12108, SPARK-3639, SPARK-3638

Orangut’s flagship learning management product for schools and colleges.

- Architect and senior developer
- Designed and implemented the product

Tech stack: Java/JEE, Spring, Hibernate, Tomcat

SpamKiller Android application is aimed at filtering spam SMS messages in Android devices (2.2+).

- Implemented the application just out of frustration caused by SMS spammers. Had put it up on Google Play store for free (without ads).