-
AllyAI – Sharepoint Connector – Authentication
Ally AI – Sharepoint Connectors – Authentication Problem #1 : How Do I Authenticate With Sharepoint ? def load_credentials(self, credentials: dict[str, Any]) -> dict[str, Any] | None: self._credential_json = credentials auth_method = credentials.get( "authentication_method", SharepointAuthMethod.CLIENT_SECRET.value ) sp_client_id = credentials.get("sp_client_id") sp_client_secret = credentials.get("sp_client_secret") sp_directory_id = credentials.get("sp_directory_id") sp_private_key = credentials.get("sp_private_key") sp_certificate_password = credentials.get("sp_certificate_password") if not sp_client_id: raise ConnectorValidationError("Client ID is required") if not sp_directory_id: raise ConnectorValidationError("Directory (tenant) ID is required") authority_url = f"{self.authority_host}/{sp_directory_id}" if auth_method == SharepointAuthMethod.CERTIFICATE.value: logger.info("Using certificate authentication") if not sp_private_key or not sp_certificate_password: raise ConnectorValidationError( "Private key and certificate password are required for certificate authentication" ) pfx_data = base64.b64decode(sp_private_key)
-
Interview – Smart Search Engine
Smart Search Engine One of the business problems I worked on at Harman was improving product discovery through Smart Search on Harman.com The challenge was that customers don’t always search using the exact terminology used in the product catalog. For example, a customer may search using a natural description of what they need, while the catalog contains product names and structured attributes. A traditional keyword search can therefore return irrelevant or incomplete results. The objective of Smart Search was to understand the user’s intent better and surface more relevant products.
-
Interview – ML Model Use Cases
ML Model Use Cases In my Travellopia work, I worked across different stages of the customer lifecycle. First, for new leads, we predicted which leads were more likely to convert so that the sales and marketing teams could prioritize them. Second, for existing customers, we predicted the probability of rebooking within 0–60 days and identified their preferred destinations so that we could provide personalized travel packages. Third, we used browsing behavior such as pages visited, quote requests and email signups to identify customers whose booking propensity was increasing. Finally, we recommended ancillary products to customers because increasing ancillary attachment directly
-
Ally AI – Connectors – Sharepoint
Ally AI – Connectors – Sharepoint Page-1: Explanation: Page-2: Explanation: Page-3: Explanation: Page-4: Explanation: Page-5: Explanation: Page-6: Explanation: Page-7: Explanation: Page-8: Explanation: Page-9: Explanation: Page-10: Explanation: Page-11: Explanation: Page-12: Explanation:
-
Systems Thinking – The Loop Law
-
Systems Thinking – The Delay Lay
-
Systems Thinking – The Ripple Law
-
Systems Thinking – The Iceberg Law
-
Generative AI Beyond Text – Essential Tools
Generative AI Beyond Text – Essential Tools
-
Safety & Guardrails – Essential Tools
Safety & Guardrails – Essential Tools
