AI Workflow Implementation Contractor for Partner Operations (Crossbeam MCP / Claude)

We are building a portfolio of AI-enabled partner-operations workflows for a go-to-market organization. The immediate work includes evaluating and implementing account-mapping workflows using Crossbeam MCP, designing a Claude-based deal-management assistant or skill, and improving manual partner eligibility and onboarding processes. This is not a website-design role, a front-end development role, or a generic prompt-engineering engagement. We need a senior, hands-on builder who can discover the current process, define the solution architecture, integrate business systems safely, implement and test the workflow, and leave the internal team with durable documentation and a supportable operating model. The work will be organized as a sequential project queue. The current planning target is approximately four projects for this contractor, delivered one at a time rather than simultaneously. The exact sequence will be finalized during discovery. Initial project themes Determine where Crossbeam MCP adds practical value to partner account mapping, prioritization, and deal workflows. Design or implement a Claude-based deal-management assistant or reusable skill that surfaces the right partner context at the right point in the deal process. Map and improve partner eligibility and onboarding workflows that currently depend on manual coordination. Identify additional high-value partner workflows, rank them by impact and implementation risk, and convert the approved items into a sequenced delivery plan. These are confirmed themes, not a complete technical specification. The contractor's first responsibility is to turn them into implementation-ready requirements. What you'll do Lead structured discovery with partner leadership, Revenue AI, go-to-market operations, and the people who perform the current work. Document the current process, system owners, data sources, permissions, manual handoffs, decision rules, exceptions, and failure points. Produce clear current-state and future-state architecture or workflow diagrams that technical and nontechnical stakeholders can validate. Define the first project's scope, acceptance criteria, dependencies, risks, and implementation plan before build begins. Evaluate the available Crossbeam MCP capabilities against the actual account-mapping and deal-management use cases. Design and implement AI-assisted workflows using Claude, MCP, APIs, webhooks, and the organization's existing business systems where appropriate. Define how account and partner context is retrieved, filtered, presented, acted on, and written back without exposing unauthorized data. Build appropriate human review, approval, escalation, and fallback paths for actions that should not be fully autonomous. Implement authentication, permissions, structured inputs and outputs, retries, timeout handling, logging, and useful error notifications. Create test data and test cases for expected paths, missing data, conflicting records, ambiguous matches, stale information, permission failures, and downstream outages. Support stakeholder validation, pilot rollout, production release, and an agreed post-launch stabilization period. Produce durable documentation: solution architecture, workflow diagrams, integration maps, decision rules, test plan, runbook, known limitations, and knowledge-transfer materials. Required experience Personally designed and shipped production business-process automations involving multiple systems, APIs, or data sources. Hands-on API integration experience, including REST APIs, JSON, webhooks, authentication, pagination, rate limits, retries, timeouts, and failure handling. Experience implementing Claude, another production LLM, or an AI agent inside an operational workflow rather than only building a standalone chatbot. Ability to translate ambiguous manual processes into explicit rules, states, exceptions, ownership, and measurable acceptance criteria. Experience working with CRM, sales operations, Revenue Operations, deal-desk, partner operations, or comparable go-to-market workflows. Practical understanding of data permissions, least-privilege access, auditability, and human approval boundaries. Ability to lead discovery with nontechnical stakeholders and work directly with internal technical owners. Strong written documentation, diagramming, testing, and knowledge-transfer habits. Nice to have Direct Crossbeam implementation or administration experience. Experience with account mapping, partner overlaps, partner-sourced or partner-influenced pipeline, co-selling, deal registration, partner eligibility, or partner onboarding. Model Context Protocol implementation experience. Experience building Claude skills, tools, or agent workflows. CRM automation and data synchronization experience. Evaluation frameworks for measuring workflow accuracy, adoption, time saved, exception rate, and business impact. Working model and access You will be the implementation lead for the assigned workflow, partnering with internal go-to-market and Revenue AI stakeholders. We expect to provide stakeholder access, guided visibility into relevant systems during discovery, and approved test data where available. All implementation work must be performed through Zoom or an equivalent live screen-sharing session. An authorized internal team member will operate the Crossbeam, Claude, CRM, and other business-system interfaces under the contractor's direction. Direct contractor credentials will not be provided. You must be comfortable diagnosing configuration in real time, directing the internal operator through exact changes, validating each result on screen, and maintaining a written record of the settings changed and evidence observed. Work that cannot be completed safely through this model must be identified during discovery before it is included in the approved implementation plan. The current CRM, automation platform, cloud environment, data warehouse, and deployment process will be confirmed during discovery. Please do not assume a specific stack in your proposal. Expected deliverables for the first project Discovery summary and prioritized problem statement. Current-state workflow and system map. Future-state architecture with data flows, permissions, human checkpoints, and failure paths. Written scope, acceptance criteria, implementation plan, and risk register. Working pilot or production implementation for the approved use case. Test evidence covering normal, exception, permission, and dependency-failure scenarios. Deployment, rollback, monitoring, and support plan. Runbook, architecture documentation, and knowledge-transfer session. What success looks like Stakeholders agree on the workflow, ownership boundaries, data sources, and acceptance criteria before implementation begins. The delivered workflow retrieves and presents the correct partner and deal context for the approved use case. Data access follows documented sharing and permission rules. Ambiguous matches, missing information, unauthorized requests, and dependency failures produce safe, understandable outcomes. Human review is preserved wherever the business has not approved autonomous action. The workflow is tested, observable, documented, and maintainable by the internal team. The first project creates reusable patterns for the remaining sequential projects.