Everything posted by Mona Dhaliwal
-
How Can Prompt Design Influence the Quality of AI Decisions?
During the CAISA training from Benchmark , I got introduced with the term "PROMPT" which is a basic requirement during the designing of the AI solution. The instructions included in the prompt defines the workflow and cycle of the AI solution. Incomplete steps or any Gap in the prompt may lead to non-functional model or inconsistent results to the queries. During the practical hands on working on an AI solution, it was evident that on changing the prompt instructions the results/outcome were varied and also the required outcomes were easily achievable. Prompt design is the heart or the backbone of any AI solution
-
How Do Roles Change When AI Becomes Part of the Team?
The most recent example of an AI solution was dustbin based on AI segregating the waste as per bio degradable or non degradable. This application of AI is a commendable shift in the work being performed by a human resource vs the machine + AI solution. This is technology wherein the human efforts and exposure to waste is considerably reduced. Also it has improved the efficiency of work. In my scenario these type of jobs wherein human life may be at risk to exposure to gases, chemicals or involving risk to human life etc should be taken over by Artificial intelligence, rather than being dependent on a human resource.
-
How Should an AI-Infused Process Be Audited?
First of the the AI infused process should be audited for its customer centric solution as to if it is able to provide the solution to the customer requirements without human interventions. Second part of the audit should be the scenarios which are unpredictable and can lead to incomplete cycle or solution provision for which the AI has been created. The most challenging part is to draft or provide solution for all the possible scenarios and later on it can be identified during the Audit process.
-
Change Management
DMAIC and change management go hand in hand. Both the activities are interlinked. The Project team may define all the phases of DMAIC , however without the actual implementation and acceptance of the strategy the six sigma assessment and completion will remain incomplete. Define -Awareness towards the change Measure & Analyze-Ensure the change to be implemented Improve-Provide appropriate training on the required changes. Control-Redefine the changes and its control parameters. MBB should be able to utilize the resources and increase awareness towards the implementation and benefits of DMAIC as a long term risk assessment and solution in place of following the old standard approaches. Involving the team in each step and identify the stakeholders for the DMAIC. The Assay of an active in the finished drug product was consistently achieved for about 40-50 batches , however suddenly in 3-4 batches a lower trend was observed although within the limits. On application of six sigma the lower PPK value resulted into orange zone. On investigation the root cause for the lower assay data was linked to the change in the input API PSD at d10 values. On redifining the controls the Assay of the FP was again achieved and the tighter controls were defined for the input API PSD.
-
Control Phase
First of all a detailed process Map should be prepared during the design phase followed by brainstorming or risk assessment of the defined process. The risk assessment should be able to identity the low, medium and high risks processes. To be able to design an efficient process /system a thorough risk assessment is essential to achieve the smooth functioning and avoid re-working.
-
Key Risk Indicators (KRIs)
Key risk indicators is an essential measurable metric for risk assessment of the products and process. KRI should be identified and controlled to have a sustainable product and process and maintain the company's brand. Key performance indicators measure the Business objectives and projected sales etc. Essentially Key risk indicators provide clear and concise way of guiding the high risk parameters to the bench level workers as required for a harmonized process understanding. One of the disadvantage for a KRI may be to identity a measurable KRI and define a working limit/control strategy.
-
BPR vs Lean Six Sigma
Business Process Reengineering (BPR) and Lean Six Sigma (LSS) are methodologies aimed at improving business processes, but they differ in their approaches and scope. BPR and Lean Six Sigma are used to enhance efficiency, reduce costs, and improve quality. BPR is used when processes are not streamlined and need to be reassessed or need lean /waste reduction. Whereas LSS is used when processes need incremental improvement, or variation reduction, or least defects. Both can be applied simultaneously depending on the process robustness assessment.
-
What can make an AI Agent a Joy to Use?
Making AI more interactive and creating an awesome customer experience first priority is always to be able to provide the genuine and satisfactory answers/solution with limited questions. The queries asked should be crisp so that the customer is able to understand and respond and reaches to the solution part quickly .Unnecessary of drama or playfulness may lead to customer no response or rejection. The features to enhance the customer experience may be incorporated such as emoji's or stickers. These can be included to make a customer friendly welcome message and with a feed of actual human voice or modulations which can change with the input sentiments/variables. Making it a wonderful interaction. Similarly the rewards system/point based system can be included in case of a lengthy questionnaire in order to motivate the customer for providing the information. In case of elderly people interaction with the AI solution the system should be customized for the Geriatric population which can include local language voice based and text in local language. Because most of elderly population are not aware of English language and also not aware of the technology. There should be light/illumination feature included in the AI solution which can be used during no background electricity and also bigger font for the elderly population to make it more customer friendly.
-
Design Your Dream AI Agent for the Future
Currently the Review of documents for the adequacy and accuracy completeness is critical for the product development. In the current scenario the review is manually performed by the reporting manager. Generally it is anticipated that approximately 80% of accuracy related corrections are identified within the first review ,however it may vary from person to person review process. I would want to build an AI solution which is capable to identify the major and minor errors and suggest corrections to the Author. In this case the AI agent needs a vigorous training on the type of errors that can be encountered by providing details about the error types and corrections to be done. If the note book's or the Electronic note book are reviewed by the AI agent it would harmonize the review process and also it may lead to identity 99% of the errors and suggest respective corrections. Even the Adequacy feature can be built in the AI solution to review the experiments for completeness and data generation required and conclusions. This would help in taking it to next level in the review process. Building an AI solution towards the document review would lead to improved & Harmonized process with error free documents.
-
Can AI Make the “Right” Call in an Ethical Dilemma?
Scenario -The BPO manages to provide essential commodities (Grains, pulses, oil, condiments) for the preparation of canteen food at the clients main office , with an employee strength of 500. In order to improve the profit margin the BPO service provider changes the vendor for the essential commodities without the quality check requirements. Due to this change, the quality of the food gets affected and to some extent the taste quality is also not maintained. The Reimbursement of payment though the AI solution for the grains ,pulses etc. is done basis the Invoice uploaded by the BPO. The bill is reimbursed by the name of BPO and not the original vendor/supplier. In the above situation AI solution will not be able to reject the poor quality of grains untill each vendor is registered and some provision of the quality checks /variable included in the program. The AI will continue to re-imburse the bill payments if the variable for the quality check is not included in the program. AI Solution program should be upgraded to include the quality check certificate from an authorized government agency and also to register the vendor /suppliers for the essential commodities.