AI Solutions Architect

Required:

• 3+ years of experience building and shipping internal systems, workflow automation, or AI-enabled tools in real organizational settings, with demonstrated recent experience deploying LLM-enabled workflows, RAG systems, agentic automation, or AI assistants that real users actually use.
  
• Strong experience integrating productivity, CRM, and knowledge systems such as Google Workspace, Airtable, Slack, shared drives, email, CRMs, or document repositories, and a preference for designing on top of existing infrastructure rather than defaulting to custom builds.
  
• Strong process and workflow design skills, including the ability to map how work actually happens, identify friction points, and design human-in-the-loop systems with approval gates, escalation paths, audit trails, and override mechanisms.
  
• Demonstrated ability to drive adoption among non-technical users through discovery sessions, training, documentation, feedback loops, and iterative improvement.
  
• Sound data-handling judgment, including PII awareness, access control, source traceability, confidentiality, and responsible use of third-party AI tools.
  
• Strong written communication skills, including the ability to document technical designs, write SOPs, and produce handoff materials for non-technical stakeholders. A pragmatic, build-and-ship orientation: comfort moving from ambiguity to working systems quickly, and willingness to sit with end users to understand their work before designing for it.
  

Preferred Qualifications

• Experience working in or with donor-funded environments such as nonprofit, international development, journalism, media support, or civic technology, including familiarity with proposal operations, bid/no-bid workflows, donor compliance, capture planning, and partner coordination. Experience with workflow automation platforms such as n8n, Zapier, Make, Airtable automations, or comparable tools.
  
• Experience with RAG systems, vector databases, document retrieval, metadata design, or knowledge-base maintenance, particularly in contexts where keeping knowledge current is a design constraint.
  
• Experience with document parsing of PDFs, DOCX files, donor solicitations, compliance matrices, and other unstructured content.
  
• Experience with prompt evaluation, AI quality assurance, model testing, red-teaming, hallucination monitoring, or structured output validation.
  
• Experience designing CRM enrichment, relationship intelligence, partner mapping, donor engagement tracking, or stakeholder mapping, and working knowledge of API hygiene, scoped credentials, OAuth flows, secrets management, and vendor security review.
  

Opportunity Fit Evaluator

• Inputs: Donor portals, newsletters, mailing lists, partner forwards, procurement notices, and other opportunity sources, using APIs where available and scheduled ingestion or lightweight parsing where APIs are not available.
  
• Behavior: Scores each opportunity against a defined BD rubric, including organizational capabilities, thematic fit, geographic focus, donor eligibility, compliance requirements, risk level, and strategic value. Confirms hard eligibility gates early and flags weak-fit, ineligible, or high-risk opportunities before staff invest significant time.
  
• Output: A ranked opportunity queue with rationale, confidence signals, eligibility notes, risk flags, and suggested next steps. The system should also surface relevant past proposals, technical experience, consortium partners, sub-awardees, donor history, and relationship context from the CRM and knowledge base.
  
• Human gate: BD reviewers accept, reject, escalate, or request more information before an opportunity advances to capture planning.
  

Application Project Manager

• Inputs: Donor solicitations, RFPs, NOFOs, application guidelines, eligibility documents, compliance instructions, budget templates, annex requirements, and submission portals.
  
• Behavior: Parses solicitation materials and extracts required deliverables, attachments, deadlines, compliance requirements, formatting rules, submission steps, budget requirements, and review dependencies. Converts these into a structured work plan, proposal calendar, and responsibility matrix.
  
• Output: A dynamic application tracker with tasks, owners, deadlines, dependencies, review checkpoints, and submission requirements, usable in the BD team’s preferred workflow tools, such as Airtable, Slack, email, or an approved lightweight review interface.
  
• Human gate: BD leadership or assigned proposal leads review and approve the work plan before assignments, reminders, or downstream automations are triggered.
  

Proposal Scaffolding Tool

• Inputs: Donor priorities, solicitation documents, concept notes, prior proposals, capability statements, approved technical language, partner profiles, lessons learned, program examples, evaluation criteria, and relevant institutional knowledge.
  
• Behavior: Retrieves relevant knowledge assets and generates early-stage proposal scaffolds, including concept frameworks, narrative outlines, partner discussion prompts, donor-aligned talking points, compliance-aware section prompts, and draft review questions. Supports staged critique aligned to donor priorities, review criteria, and compliance expectations.
  
• Output: Structured proposal-starting materials that help BD and program staff move faster from opportunity assessment to concept development. Outputs should include source references where possible and distinguish between retrieved institutional material, AI-generated synthesis, and areas requiring human judgment.
  
• Human gate: BD and technical leads review, revise, approve, or reject generated scaffolds before they are used in partner discussions or proposal drafting.
  

CRM Relationship Intelligence Layer

• Inputs: Existing CRM or Airtable records, donor and partner contact data, engagement notes, prior opportunities, proposal history, event participation, conference intelligence, relationship ownership information, and approved external or internal data sources.
  
• Behavior: Structures and enriches relationship intelligence so BD staff can understand donors, partners, contacts, decision pathways, and engagement history. Captures relationship strength, our strategic role with each donor, prior collaboration history, pathway mapping to decision-makers, consortium partner and sub-awardee fit signals, event and conference attendance intelligence, pre-RFP signals, Terms of Reference shaping opportunities, and ownership rules for data maintenance.
  
• Output: An Airtable-anchored or CRM-connected relationship intelligence layer that surfaces useful context during opportunity screening, capture planning, partner outreach, and proposal development. The system should make relationship data easier to maintain, not harder.
  
• Human gate: Staff review and confirm sensitive relationship notes, ownership fields, inferred connections, and strategic recommendations before they are treated as reliable CRM intelligence.