Create
cancel
Showing results for 
Search instead for 
Did you mean: 
Sign up Log in
Celebration

Earn badges and make progress

You're on your way to the next level! Join the Kudos program to earn points and save your progress.

Deleted user Avatar
Deleted user

Level 1: Seed

25 / 150 points

Next: Root

Avatar

1 badge earned

Collect

Participate in fun challenges

Challenges come and go, but your rewards stay with you. Do more to earn more!

Challenges
Coins

Gift kudos to your peers

What goes around comes around! Share the love by gifting kudos to your peers.

Recognition
Ribbon

Rise up in the ranks

Keep earning points to reach the top of the leaderboard. It resets every quarter so you always have a chance!

Leaderboard

Come for the products,
stay for the community

The Atlassian Community can help you and your team get more value out of Atlassian products and practices.

Atlassian Community about banner
4,554,411
Community Members
 
Community Events
184
Community Groups

Enhancing Email&Tasks: Jira Cloud for Gmail with GPT-3

Email&Tasks: Jira Cloud for Gmail is a suite of three add-ons (Atlassian Marketplace, Google Workspace Marketplace connector for Gmail, and Chrome browser extension).  The add-on is designed to optimize task management processes, such as quick access to tasks and any operations with them, as well as detecting tasks in emails in an inbox, etc. We described its work in more detail in the articles - Jira & Gmail – the perfect tandem for your team's productivity, Convert emails into Jira issues and not only.

The main function around which the app is built is the conversion of emails into tasks in Jira. It is possible in manual mode and with the revolutionary 1-CLICK TASK option. This feature is made possible by an AI solution.

The team worked hard to improve the automatic task detection in emails and implemented GPT-3 technology in the app. So, let's talk about why it was done, what preceded it, and what we got as a result.

ezgif-1-1c3227ff8f.gif

What is GPT-3?

GPT-3 (Generative Pre-trained Transformer 3) is a natural language processing model developed by OpenAI. It is the latest and most powerful version of the GPT series of models based on a deep-learning architecture called a transformer. GPT-3 contains 175 billion parameters.

GPT-3 has many capabilities, including translation, question answering, completion, and generation. It can understand natural language and generate human-like responses, making it useful for tasks such as chatbots, virtual assistants, and speech applications. GPT-3 can also be customized to perform specific tasks such as sentiment analysis, generalization, and language modeling.

One of the unique features of the GPT-3 is its ability to perform multi-shot and zero-shot learning. Multiple-shot learning involves training the model on only a few task examples. In contrast, zero-shot learning allows the model to perform tasks for which it has not been specifically trained using existing knowledge and context.

2023-03-16_15-14-25.png

Implementing GPT-3 in the Email Tasks Jira Cloud app for Gmail

Previously, our developers used the Codeq API to recognize tasks in the email content. Codeq's proprietary NLP technology offers advanced deep-learning models and linguistic analysis tools to extract rich representations from your textual data. Codeq API is optimized for accuracy and precision in understanding and responding to task classification. Despite all its capabilities, this solution did not fully satisfy us, so we decided to implement GPT-3.

What is the difference? Let's try to compare these two technologies.

Feature

Codeq API

GPT-3

Purpose

Natural language is recognized, but the result is often negative in terms of quality

Wide range of natural language processing tasks

Accuracy

High

Very high accuracy in natural language recognition. Much better than Codeq API

Model architecture

Deep learning

Transformer-based architecture

Performance

Highly specialized in task classification 

Broad-based but may require fine-tuning for specific tasks

Language Support

English

Supports a wide range of natural languages

API availability

Available for commercial use

Available for commercial use

Cost

Paid subscription

Paid subscription

Training data

Labeled text

Large-scale general-purpose text corpora

Since the application works with natural language, switching to this technology was logical. Let's look at an example of email detection with GPT-3 already implemented.

Use cases and benefits of GPT-3 in the Email&Tasks: Jira Cloud for Gmail app

This technology allows you to recognize the text of emails and transform them into Jira tasks. 

How does it work?

  1. Install the app from the Atlassian Marketplace and its connector from the Google Workspace Marketplace.
  2. Open the email you want to convert to a task and click the Email&Tasks icon.
    1.png
  3. Select the TRY 1-CLICK TASK (AI-BASED) option.
    2.png
  4. And that's it. The app recognizes your email and offers a draft of the upcoming task. It also acknowledges the priority from the text and automatically assigns the email recipient as Assignee.
    3.png

Now you just need to click the Create issue button.

So, a few seconds and you're done. Note that if the email contains attachments, checking the Include email Attachments box will automatically attach them to the issue.

Takeaway

Implementing GPT-3 in Email&Tasks: Jira Cloud for Gmail app is not the only thing we are preparing to improve its performance. Everything is for the user's convenience; everything is for optimizing workflows. Try to test this AI-powered app, and be sure to write to us about your impressions. We've got a 30-day trial for you to see its benefits. Also, note that the add-on is free for teams of up to 10 users. Enjoy!

0 comments

Comment

Log in or Sign up to comment
TAGS
AUG Leaders

Atlassian Community Events