Skip to main content

7 posts tagged with "sprint"

View All Tags

· 3 min read

Work in Progress​

  • The team continue using Trello boards to track the cards in progress and using Slack to communicate blockers.
  • Today the team continued working and made progress on the following streams
    1. Multilingual capability:
      • The team observed that it was tricky to have a single collection in WD with two different languages, as WD can treat collection only in a single language. To solve this challenge, the team created two different collections, one with english and other with spanish both are the web crawl data.
      • Another challenge the team faced was dynamically figuring out which collection to be pointed to fetch results depending on question asked. Ideally if question is asked in spanish we need to fetch it from spanish collection in WD. To handle this challenge, the team developed the following prompt and gave instructions to the llama model to take care of mixed language and answer cleverly.
        • You need to deal with two languages that are English and Spanish. If the question is asked in English, then please answer in English. If the question is asked in Spanish, please answer back in spanish. Context for the question can be in Spanish or English or both the languages.
      • The team is also evaluating setting the first action to prompts the user to input their desired language
    2. FAQs to Actions: the team continue adding FAQs as actions. The next steps would be leveraging the watson assistant v2 api to create a python utility that has the following flow automated
      • set up the Assistant
      • transfer actions from one assitant to another assistant
      • WatsonX FAQs to actions
    3. High value journeys: the team finished developing the following actions to walk through users through a multi-step action
      • account opening
      • i lost a debit card
    4. Setup and integration of Watsonx Assistant and Genesys: The team continue the integration of Genesys to Watson Assistant, referring Integrating with Genesys. At around 12:30pm the team developed a static website hosted on COS that successfully integrate Watson Assistant with Genesys.
    5. Conversational Search Capability: For the questions that are not covered, GenAI conduct a search through the document collection. We tested token amounts to ensure comprehensiveness and conciseness of response
      • currently we are using the meta-llama/llama-2-70b-chat model, with max_new_token 300, repetition_penalty 1.1, and temperature 0.
    6. Website development using the Lendyr codebase: the team plan to integrate our Watson Assistant with the Lendyr codebase and customize it to match our client's information. This will provide the capabilities such as carousel, Map iframe, Journey, Email deep link, Mortgage Calculator, on top of the LLM Search.
      • at around 6pm, the team got the integration between the Lendyr codebase to work with Genesys integration

Currently Tracking​

  • The team is working on pushing solution documentation to GitHub.

Next Steps​

  • The team will continue to work through the use cases and demos.

· 2 min read

Work in Progress​

  • The team setup Trello boards to track the cards in progress.
  • Today the team continued working and made progress on the following streams
    1. Multilingual capability:
      • Fine tune FAQs as the answers from the Conversational Search could response perfectly around 35%, but for some questions the model does not perform well due to lack of Spanish sources, and sentences not cutting off properly.
      • The team continue testing prompts, adding spanish documents to Watson Discovery, and adding actions to Watsonx Assistant.
    2. FAQs to Actions: the team continue adding FAQs as actions.
    3. High value journeys: the team continue developing personas and journeys for the questions that would require human agents. We developed actions to walk through users through a multi-step journey.
    4. Setup and integration of Watsonx Assistant and Genesys. The team continue testing the integration of Genesys to Watson Assistant.
    5. Conversational Search Capability: For the questions that are not covered, GenAI conduct a search through the document collection. We tested token amounts to ensure comprehensiveness and conciseness of response
      • currently we are using the meta-llama/llama-2-70b-chat model, with max_new_token 300, repetition_penalty 1.1, and temperature 0.
    6. Website development using the Lendyr codebase: the team plan to integrate our Watson Assistant with the Lendyr codebase and customize it to match our client's information. This will provide the capabilities such as carousel, Map iframe, Journey, Email deep link, Mortgage Calculator, on top of the LLM Search.

Currently Tracking​

  • The team is working on pushing solution documentation to GitHub.

Next Steps​

  • The team will continue to work through the use cases and demos.

· 2 min read

Work in Progress​

  • Today the team setup the Watsonx Assistant and the integration with Watson Discovery and Watonsx.ai. At around 10am we had the first version of WA + WD + WX up and running. We tested the Conversational Search capabilties on three questions and it worked well in terms of relevance and accuracy.
  • The team then worked in the following streams:
    1. Multilingual capability:
      • Ensure Spanish websites are added to the Watson Discovery collection
      • Through prompt engineering, we develop the multilingual capability that has the generative AI capability to answer previously unseen questions
    2. FAQs to Actions: for the questions that are frequently asked, we developed actions to ensure standard responses, which also optimized the response time.
    3. High value journeys: for the questions that used to require human agents e.g. Can you help me open an account?,I lost my debit card, we developed actions to walk through users through a multi-step journey
    4. setup and integration of Watsonx Assistant and Genesys.
    5. Conversatiol Search Capability: For the questions that are not covered, GenAI conduct a search through the document collection. We tested token amounts to ensure comprehensiveness and conciseness of response; currently we set max_new_token as 200

Currently Tracking​

  • The team is working on pushing solution documentation to GitHub.

Next Steps​

  • The team will continue to work through the use cases and demos.

· One min read

Work in Progress​

  • Today the team worked on setting up an environment and plotting the usecases and personas.
  • The team listed potential FAQs that we plan to answer using the proposed solution.
  • We listed the instances we need for the upcoming project as listed below
    • Watsonx Assistant
    • Watson Discovery Plus
    • Watson Machine Learning (essentials or standard)
    • Cloud Object Storage: Needed for watsonx platform
    • DB2: needed for storing watson assistant logs for analytics
    • Cloud pak for Data platform
    • WatsonX platform
    • Watson Studio: needed for data visualizations and integrate data sources with cogos
    • Cognos Dashboard Embedded: to visualize logs and analytics
    • Watson Knowledge Catalog: needed for data visualizations and integrate data sources with cogos
    • Watson NLU: multilingual understanding
  • The team tried provisioning an environment from Techzone but the provisioning failed.
  • The team received access to an IBM Cloud environment at around 6pm EST.

Currently Tracking​

  • The team is working on pushing solution documentation to GitHub.

Next Steps​

  • The team will continue to work through the use cases and demos.

· One min read

Work in Progress​

  • Continue working on the task to calculate which % of IT helpdesk requests can be handled by our proposed solution.
  • We developed the end-to-end pipeline with the following components and tested on 1000 example data
    • Keywords extraction using WatsonX models
      • IT helpdesk requests
      • knowledgebase resources
    • Question Generation using WatsonX
      • generate questions that the knowledgebase could answer for
    • Similarity Analysis
      • Leveraged spacy packages to tokenize keywords and calculate the cosine similarity score between each IT helpdesk requests and all knowledgebase keywords

Currently Tracking​

  • The team is working on pushing solution documentation to GitHub.

Next Steps​

  • The team will continue to work through the use cases and scale the usecase

· One min read

Work in Progress​

  • Client shared sample data to the team and asked if we could estimate which % of IT helpdesk requests can be handled by our proposed solution.
  • We plan to develop the end-to-end pipeline using a sample of datasets.
    • apply watsonx models to extract keywords from IT helpdesk requests.
    • apply watsonx models to extract keywords from knowledgebase resources.
    • apply watsonx models to generate questions that the knowledgebase could answer for.
    • apply similarity analysis between each IT helpdesk requests keywords to the complete corpus of knowledge-base keywords and calculated the cosine similarity score between each incident and all knowledgebase keywords

Currently Tracking​

  • The team is working on pushing solution documentation to GitHub.

Next Steps​

  • The team will continue to work through the use cases and scale afterwards

· One min read

Work in Progress​

  • Today the team discussed how to efficiently setup the Watsonx.ai extension and Watson Discovery extension.
  • The team then walked through how we setup prompt engineering within Watsonx Assistant.
  • The team then walked through a demo of the setup and integration of Watsonx Assistant and ServiceNow.

Currently Tracking​

  • The team is working on pushing solution documentation to GitHub.

Next Steps​

  • The team will continue to work through the use cases and demos.