Tayyab Rashid

I designed and created a management system for solar power plants, successfully implementing a multi-tenant SaaS solution. In addition, I successfully implemented the onboarding of new clients and security.
I was in charge of everything, working as an architect and team lead and interviewing new candidates.

This application is based on Azure technology. I implemented a multi-tenant monitoring system, inventory management system, incident system, and alarm system. This system is module-based.
The technologies used are Azure IoT Hub, Data Explorer, Functions, App Service, SQL Server, Application Insights, ADB2C, and Digital Twin. Also used Docker container, React, API management, identity management (IdM), and CDN service. All the infrastructure is deployed by the code that I wrote.

I implemented a Dynamics 365 solution for the new home sales department, OBOS.

Project Features:
• Dynamics 365
• Plugins
• Workflow
• JavaScript
• SMS solution
• Dynamic Marketing
• Integration with an external system
• Azure API management
• Azure Service Bus
• ASP.NET website
• Azure storage
• DevOps

Client: VVS Group Norway is located in Oslo, Norway, and is part of the plumbing and HVAC contractors industry. The company has 37 employees at this location and generates $5.04 million in sales (USD). I implemented the integration between Dynamics 365 and 24SevenOffice (ERP). All of the clients that were created in Dynamics 365 should also be duplicated within 24SevenOffice.

In this project, my main task was to integrate Infosoft with Dynamics 365. It required that all of Infosoft's clients should also be duplicated within Dynamics 365, as well as their subscriptions. Another complication was that Infosoft had various events that people could subscribe to.

CGI had developed a standard solution for membership organizations. My main task was to implement the solution for different clients and customize it to their needs.

My work on this project ran the gamut from creating different integrations to writing workflows and plugins in Dynamics 365. I also built the billing system in Dynamics 365 and integrated it with Dynamics Business Central.

A junior consultant and I created a portal solution for asylum seekers in Norway. I was responsible for the design and implementation. This included everything from creating the API, the Dynamics 365 implementation, security design, and building the front-end application with React.

Client: Fremtind was created on January 1st, 2019, as a new insurance company that came out of a merger between two of Norway’s largest financial service groups. The aim was to become the most preferable of all competitors, and position Fremtind as the life insurance company for the future, focusing on digital innovation rooted in customer insights and needs. The company needed a visual identity to help reach its goals. Fremtind is today among the most popular places to work and has marked its territory in the Norwegian market.

Project:
This was a Dynamics on-premise solution, and my main task was to integrate the Dynamics 365 solution with the core banking system for the whole group. I also created the monitoring app for the MSMQ using Angular (TypeScript). In addition to writing various workflows, plugins, and JavaScript components.

I handled the integration between Dynamics 365 and 24SevenOffice (ERP). It required that all of Dynamics 365 clients were also duplicated in 24SevenOffice. I also implemented a billing solution for the client and ensured that all the orders were created in Dynamics 365 and synced to 24SevenOffice.

Client: BKK is in the industry of renewables and the environment.

I was responsible for the data migration using SSIS which included their plugins, workflow, and JavaScript.

The client had an on-premises Dynamics 2015 solution, which I helped migrate to Dynamics 365. I built daily SSIS integration packages using KingswaySoft to synchronize banking system data with Dynamics 365, processing 10,000+ customer records with 99.9% reliability. I also spearheaded the migration of SSRS into Power BI and established AppDynamics for the client.

Client: Sparebank Helgeland is a Norwegian savings bank, headquartered in Mosjøen, Norway.

I was responsible for the entire data integration with the external system; this was handled through SSIS. The client had an on-premise solution which was called Sparebank Sogn og Fjordane.

I was responsible for the entire implementation of this Dynamics 2016 on-premise and SharePoint integration for a Norwegian insurance company.

I have recently been hired by two consultancy companies to help them with their clients. As a functional and technical consultant, I have delivered various solutions for SharePoint and Power Platform (Canvas and model-driven platform).

I built an end-to-end algorithmic trading system that automatically parses trading signals from Telegram, validates risk parameters, and executes orders across multiple MetaTrader 5 accounts with AI-optimized execution timing.

Developed the complete solution as full-stack developer:

• Back end (Python and FastAPI): Signal ingestion using Telethon SDK, classification of trade types (ENTRY, MODIFY, and CLOSE), risk validation engine, direct MT5 integration via MetaTrader5 SDK, symbol validation for broker compatibility, system readiness checks, background service for 24/7 operation, SQLite database with SQLAlchemy ORM.

• Front end (React and TypeScript): Real-time monitoring dashboard showing active trades and system status, trade history with filtering and search, manual execution interface, and risk parameter configuration. It was built with Vite for rapid development.

KEY ACHIEVEMENTS
• Successfully processes 1,000+ trades daily with 99.5% execution accuracy.
• AI agent reduces slippage by 40% through market condition analysis.
• Sub-second trade replication across 50+ accounts.
• Zero missed signals with continuous background processing.
• Comprehensive logging for troubleshooting and optimization.

I delivered an AI transformation for a Norwegian construction company as AI Solutions Architect and Lead Developer. I combined Microsoft 365 Copilot deployment with four custom AI agents built on Azure AI Foundry, automating tender preparation, safety compliance, invoice processing, and inventory management.

Phase 1 - Microsoft 365 Copilot: Deployed to users with construction-specific training. Achieved 92% daily adoption rate within 30 days. Integrated with existing SharePoint and Teams infrastructure.

Phase 2 - Custom AI Agents (Azure AI Foundry):
1) Tender Intelligence Agent: Analyzes tender PDFs using Azure OpenAI GPT-4, searches past winning bids, generates proposals, creates cost estimates from historical data, and checks compliance automatically.
2) Safety Compliance Agent: Monitors safety documentation, flags missing items, analyzes incident patterns for risk prediction, and generates audit reports.
3) Invoice Processing Agent: OCR extraction with Azure Document Intelligence, automatic matching to purchase orders, duplicate detection, and Business Central integration.
4) Inventory Management Agent: Predicts material needs from schedules, optimizes bulk ordering, tracks across job sites, and integrates with suppliers.

I architected an AI platform as the AI Solutions Architect and Lead Developer for a Norwegian law firm. I combined Microsoft 365 Copilot with five specialized legal AI agents to automate research, contract review, client intake, document generation, and billing intelligence.

Phase 1 - Microsoft 365 Copilot: Deployed to lawyers and admin staff with legal-specific training and Lovdata (Norwegian legal database) integration.

Phase 2 - Custom Legal AI Agents (Azure AI Foundry):
1) Legal Research Agent: RAG implementation using Azure AI Search with vector embeddings for semantic search across the Lovdata database, generates case law summaries, multi-language support (Norwegian, English, and Urdu).
2) Contract Intelligence Agent: Automated risk analysis, clause comparison against templates, identifies missing and risky terms, and Norwegian law compliance checking.
3) Client Intake Agent: Extracts info from forms and email, conflict checking, automatic case categorization, file creation, and initial communication drafting.
4) Legal Document Generator: Template-based creation, auto-fills client data, smart clause suggestions, multi-language generation.
5) Billing Intelligence Agent: Monitors activities (emails, calls, and documents).

I was selected via Toptal (top 3% of developers globally) by HUB International, North America's largest insurance broker, after they were unable to find qualified local expertise in the USA. I worked as a Dynamics 365 Integration Specialist and SharePoint Architect, specializing in enterprise-scale digital transformation, serving thousands of users across multiple countries.

PRIMARY RESPONSIBILITIES
SharePoint-Dynamics 365 Integration: Architected a solution to mirror Dynamics 365 security into SharePoint at the document level using Microsoft Graph API. SharePoint's standard access gives all site members access, so I developed a mechanism that dynamically creates folders, assigns correct permissions based on CRM ownership and security roles, and removes unauthorized sharing - all controlled from Dynamics 365. This ensures document access follows the same permission model as CRM data.

I led the development of a complete membership management system for Fagforbundet, Norway's largest trade union with 400,000+ members, as a full-stack lead developer. I replaced the legacy Fane2 solution with a modern platform built on Dynamics 365/Dataverse and Azure, which handles member registration, dues collection (including invoicing and payments), and conflict resolution.

PRIMARY RESPONSIBILITIES
Integration Architecture: Designed a service platform using Azure Service Bus for reliable messaging, Azure Functions for background processing (payment reconciliation and invoice calculation), and Logic Apps for automated workflows, all integrated with Business Central ERP for invoicing and financial management.

KEY INTEGRATIONS
Public registration form (fagforbundet.no) → Cosmos DB/Azure SQL → Functions validate → Service Bus → Dataverse. Payment processing from banks → Functions calculate discrepancies → Match invoices → Update member accounts → Business Central sync.

TESTING AND DEVOPS
Implemented comprehensive E2E tests with SpecFlow and Gherkin, unit tests for business logic, and Azure Pipelines for CI/CD automation.

I joined a startup as an integration specialist and lead developer to architect and implement an integration platform from scratch. The platform processes 100,000+ events daily from external systems and devices with secure authentication and real-time webhooks.

ARCHITECTURE AND IMPLEMENTATION
• Integration Infrastructure: Designed an event-driven architecture using Azure App Services for API hosting, Azure Functions for serverless data processing (auto-scaling based on load), and Azure Service Bus for message queuing with guaranteed delivery.
• Security Implementation: Implemented authentication and authorization for all endpoints and secure credential storage in Azure Key Vault, API key management, and OAuth2 flows for third-party integrations.
• Webhook System: Designed webhook endpoints for real-time data ingestion from third-party sources, implemented validation and async processing to prevent timeouts, added retry logic with exponential backoff for failed deliveries, and used dead-letter queues for error handling.
• Data Processing: Real-time processing pipelines for urgent data, scheduled batch processing for high-volume data, comprehensive error handling with automatic retries, and a complete audit trail for compliance.

I designed and implemented a complete integration platform for Bakke AS as a full-stack developer, synchronizing data across multiple internal and external systems, including Dynamics F&O and various business applications, and enabling Power BI analytics through Dataverse.

ARCHITECTURE AND IMPLEMENTATION
• Integration Infrastructure: Azure Service Bus for reliable messaging between systems, Azure Table Storage for high-volume audit logs, Logic Apps for workflow orchestration, Azure App Service for API hosting, Dataverse as a central integration hub, and a React monitoring portal for the operations team.
• Integration Patterns: Implemented both real-time (event-driven) and batch integration patterns, comprehensive error handling with automatic retries and dead letter queues, support for different data formats (JSON, XML, and CSV), and transformation logic for data mapping between systems.
• Monitoring Portal (React): Built a custom web portal for the operations team to monitor all integrations, view Service Bus messages and queues, track errors and failures with detailed logs, manual retry capabilities for failed messages, real-time status dashboards, search and filter functionality.

I delivered multiple successful Dynamics 365 and Power Platform projects as a consultant across various industries:
• Capgemini/Walwil: Voice of Customer (VOC) solution with Qualtrics integration, PowerApp, Power Automate, and SharePoint
• Hussoppen: Dynamics CRM 365 implementation with Oracle database integration via Azure Data Factory.
• Elkonor: SharePoint intranet solutions with custom web parts and Power Automate workflows.
• ECIT: Microsoft 365 automation solutions across Power Platform.
• VVS Gruppen: Dynamics 365 integration with 24SevenOffice ERP.
• Nemitek: Dynamics 365 integration for membership organization.

Common Technologies: Dynamics 365, Power Platform, SharePoint, Power Automate, Power Apps, Azure Functions, .NET, and JavaScript

I led an integration project for Norwegian insurance organization Hussoppen, implementing Dynamics 365 CRM and integrating with legacy Oracle database systems.

I designed and built a complete data synchronization solution using the KingswaySoft SSIS Integration Toolkit. SSIS packages were developed in Visual Studio to extract customer and subscription data from Oracle, apply transformation logic, and load the data into Dynamics 365 and Dataverse. I worked with Microsoft Fabric, including Azure Data Factory, to design the overall integration architecture. External connections to Norwegian business registers were implemented to support automated data validation.

I developed .NET back-end logic and custom Dataverse plugins for business automation. I successfully migrated all data while maintaining business continuity.

I built an AI-powered intelligence platform for a real estate investment firm that lets staff query internal company knowledge and live Microsoft 365 data through a secure, role-aware chat interface. The back end uses FastAPI with Azure AD (Entra ID) authentication and role extraction, a model-agnostic AI layer that allows the underlying LLM to be swapped without code changes, and a RAG knowledge base on ChromaDB for grounded answers from internal documents such as deal memos, market analyses, and investor FAQs. I integrated Microsoft Graph to surface live M365 data, added topic and confidence guardrails to keep responses on-domain, and built an admin control panel for configuration and interaction logging. The React/TypeScript front end uses MSAL auth, a role selector, and a live M365 data demo. It was delivered as a working, deployable proof of concept on Azure.

I architected and built a production-grade, multi-tenant B2B SaaS platform from the ground up. The application combines a real-time analytics dashboard, third-party data integrations, an integrated learning module, and automated reporting into a single enterprise product. Key engineering work included a strict multi-tenant data model with organization isolation enforced at the database level via Supabase row-level security; OAuth integrations with Microsoft 365 and Google Workspace to ingest live organizational data; a dashboard layer that aggregates that data into executive-level KPIs and charts; and a learning module (LMS) with video lessons, quizzes, certificates, and learning paths. Enterprise capabilities include SSO (SAML/OIDC), white-labeling, full bilingual (Norwegian/English) internationalization, Stripe billing in multiple currencies, and board-quality PDF report generation. It was built with type-safe end-to-end TypeScript and deployed entirely in the EU for GDPR compliance.

Stack: Next.js 14 (App Router), Supabase (PostgreSQL and Edge Functions), Stripe, Recharts, shadcn/ui, and Vercel.

I architected and built a full-stack business operations platform for a Norwegian plumbing and renovation company, replacing manual paperwork with an end-to-end digital workflow: proposals, kanban jobs, worker SMS dispatch, invoicing, and ERP sync. The hardest engineering challenge was integrating Tripletex (Norway's dominant accounting API) with session token refresh logic tied to CET midnight expiry — a time zone edge case that caused silent auth failures requiring a +2-day buffer workaround. I built a live kanban board via Supabase Realtime (REPLICA IDENTITY FULL), a public worker portal gated solely by unguessable 128-bit UUID tokens, and a server-side PDF pipeline using react-pdf/renderer. I integrated Claude Sonnet as an AI business assistant through a custom MCP server that exposes all company data as typed tools and resources, enabling natural-language querying and action across the full dataset. The monorepo shares a single Drizzle ORM schema across the Next.js app and MCP server process.

Stack: Next.js 15 App Router, TypeScript 5.7, Supabase (Postgres, Auth, Realtime, Storage), Drizzle ORM, Anthropic Claude API, MCP SDK, Twilio, Resend, Vercel.