Parag Agrawal has a total of 30 patents globally, out of which 12 have been granted. Of these 30 patents that he has filed, more than 86% patents are active. The United States of America is where Parag Agrawal has filed the maximum number of patents.
Agrawal is one of the people in the long list of Indian-origin CEOs in Silicon Valley. He has worked as a researcher in Microsoft, Yahoo! And AT&T. Later on he joined Twitter in 2011 when the company had fewer than 1000 employees and was appointed the company’s CTO in 2017 after which he took charge as the CEO in November 2021, replacing Jack Dorsey.
How Many Patents did Parag Agrawal File Every Year?
| Year of Patents Filing or Grant | Parag Agrawal Applications Filed | Parag Agrawal Patents Granted |
| 2011 | – | – |
| 2012 | – | – |
| 2013 | – | 1 |
| 2014 | 5 | – |
| 2015 | 1 | 1 |
| 2016 | 2 | 1 |
| 2017 | 3 | – |
| 2018 | 2 | 1 |
| 2019 | 5 | 1 |
| 2020 | 9 | 5 |
| 2021 | – | 1 |
| 2022 | – | 1 |
Parag Agrawal Patents
Parag Agrawal has a total of 30 patents. These patents belong to 17 unique patent families. Out of 30 patents, 26 patents are active.
How Many Patents did Parag Agrawal File in Different Countries?
What Technologies are covered by Parag Agrawal’s Patents?
One of the patents in the AI & Machine Learning category, US20190354594A1, talks about “Building and implementing persona-based language generation models,” which refers to the chatbot conversations.
A chatbot is a computer system that can be spoken to. A conversational agent, more particularly, aims to emulate human dialogue and provide a human-machine interaction based on human conversation. It can be used in text messaging applications in one case, allowing users to initiate and communicate with a conversational agent via text message. Other modalities, such as spoken language in spoken language conversational agents, are also accessible.
Conversational agents have a wide range of uses, including responding to client requests for products and services, as well as delivering customer care, technical support, education, and personal assistance.
10 Best Parag Agrawal Patents
US9117227B1 is the most popular patent in the Parag Agrawal portfolio. It has received 77 citations so far from companies like Microsoft, LinkedIn, and IBM.
Below is the list of 10 most cited patents of Parag Agrawal:
| Publication Number | Citation Count |
| US9117227B1 | 77 |
| US10650408B1 | 75 |
| US20090210429A1 | 75 |
| US20090216718A1 | 32 |
| US9892431B1 | 15 |
| US9454771B1 | 14 |
| US10248667B1 | 8 |
| US20180046918A1 | 7 |
| US10657556B1 | 6 |
| US20190354594A1 | 5 |
He has been in charge of most of the company’s ground-breaking projects at Twitter, the most important of which is the successful monetisation of the company’s advertisement-based revenue models. The ads team employed machine learning to analyse data and tailor advertisements to users under Agrawal’s leadership.
He’s also pushed to boost user growth by increasing the relevancy of the ‘Home timeline,’ rethinking the company’s technical strategy, and directing machine learning and AI across the board. In 2019, Dorsey enlisted Agrawal’s help to develop Bluesky, an ambitious Twitter-funded project to create a decentralised social media where people may moderate and promote content using their own algorithms.
Do read our analysis on one of the founders of Twitter: Jack Dorsey
List of Parag Agrawal’s Patents
| Parag Agrawal Patents | Title |
| US9454771B1 | Temporal features in a messaging platform |
| US8606771B2 | Efficient indexing of error tolerant set containment |
| US9117227B1 | Temporal features in a messaging platform |
| US9892431B1 | Temporal features in a messaging platform |
| US20180046918A1 | Aggregate Features For Machine Learning |
| US10248667B1 | Pre-filtering in a messaging platform |
| US20190354594A1 | Building and deploying persona-based language generation models |
| US10600080B1 | Overspend control in a messaging platform |
| US10650408B1 | Budget smoothing in a messaging platform |
| US10657556B1 | Click-through prediction for targeted content |
| US20200257543A1 | Aggregate Features For Machine Learning |
| US20200265101A1 | Cohort service |
| US10769677B1 | Temporal features in a messaging platform |
| US10769661B1 | Real time messaging platform |
| US20210034635A1 | Intent based second pass ranker for ranking aggregates |
| US20210097339A1 | Inference via edge label propagation in networks |
| US20210097384A1 | Deep segment personalization |
| US20210110428A1 | Click-Through Prediction for Targeted Content |
| US20210216944A1 | Deep reinforcement learning for long term rewards in an online connection network |
| US20210232590A1 | Heterogenous edges in an online network for building active online communities |
| US20210295170A1 | Removal of engagement bias in online service |
| US11157464B1 | Pre-filtering of candidate messages for message streams in a messaging platform |
| US20210350284A1 | Tackling delayed user response by modifying training data for machine-learned models |
| US11270333B1 | Click-through prediction for targeted content |
| US20220083853A1 | Recommending edges via importance aware machine learned model |
| US20220101159A1 | Recommending network connections by optimizing for two-sided implicit value of an edge |
| US20090216718A1 | System for Query Scheduling to Maximize Work Sharing |
| US20090210429A1 | System and method for asynchronous update of indexes in a distributed database |
| EP3497625A1 | Aggregate features for machine learning |
| WO2018031958A1 | Aggregate features for machine learning |