About Us

Our company is building an automation system for procurement and sales of aircraft components. The system automates the full RFQ cycle: receiving requests from the ERP, selecting suppliers, sending requests, processing replies with AI, generating quotes, and preparing commercial proposals for customers.
The solution is already operating on real RFQs and commercial data. The next step is moving from automating individual operations to a governed decision-making system that can independently handle a significant share of standard transactions and route only complex, high risk, and non standard cases to humans.

The strategic goal is to evolve Banner Copilot into an AI-supported operating and decision-making layer for Banner Aircraft International.

The platform should gradually automate standard account-management and operational workflows, support management decisions with data-driven recommendations, identify business and financial risks, and enable a leaner operating model over time.

The platform should first provide decision support and next-best-action recommendations, and progressively take autonomous actions in clearly defined, low-risk scenarios. Over time, it should help prioritize opportunities, support resource allocation, identify financing needs, assess risks around customers, suppliers, margin and cash flow, and recommend concrete management actions.

The role

We are looking for a hands-on AI Business Automation Architect who can help develop Banner Copilot not only as a procurement and sales automation platform, but as an AI-supported operating and decision-making system for the company.

The architect will work with the existing product and codebase, understand Banner’s operational and commercial processes, identify which activities and decisions can be automated, and help the engineering team bring practical solutions to production.

This is not a purely advisory role. The architect must understand the existing codebase, participate in prototyping critical components, run architectural reviews, and work with the engineering team to bring solutions to production.
The primary outcome of the role is not architecture documentation itself, but a working system that reduces manual account-management and operational work, increases transaction-processing speed, improves decision quality, identifies business and financial risks, and provides management with concrete, data-supported recommendations.

Responsibilities

• Product architecture
• Audit the current architecture
• Identify which account-management and operational functions can be fully or partially automated
• Map the key operational and management decisions currently made by employees
• Define the transition from decision support to controlled automation and limited autonomy
• Define the target architecture for the full cycle: RFQ → Quote → Sales Order → Delivery → Close → Follow-up → Management Recommendation
• Set clear boundaries between business rules, algorithms, statistical models, and AI agents
• Prepare a phased technical roadmap for evolving from the current solution to an autonomous engine
• Design a scalable, maintainable architecture to replace the current monolithic setup
• Define requirements for data, logging, observability, fault tolerance, and security
• AI and decision automation
• Design AI‑agent workflows for supplier requests, web search, and retrieval of up‑to‑date prices
• Define architecture for using LLMs, tools, confidence scoring, and human‑in‑the‑loop
• Develop exception‑handling rules, including AOG cases, high deal value, and low model confidence
• Determine where AI is truly needed versus where formal rules are more reliable
• Contribute to the development and testing of prompt logic, extraction pipelines, and agent orchestration
• Design transaction-prioritization logic based on margin, probability of closing, timing, customer risk, supplier risk and cash impact
• Define which decisions Copilot may make independently and which require human or C-level approval
• Design next-best-action recommendations for account managers and operational teams
• Establish auditability, explainability and approval controls for AI-supported decisions
• Pricing and decision‑making
• Design a Pricing Engine
• Define the architecture for calculating margin, deal probability, supplier risk, and commercial attractiveness
• Separate rule‑based, statistical, and ML approaches
• Plan the use of historical data from eMARS and other price sources
• Define a model for selecting the best price factoring recency, supplier reliability, lead times, and risk
• Management intelligence
• Design a management-recommendation layer that explains what is happening, where the risks are, what options are available, and what action is recommended
• Define regular AI-generated management reports covering sales pipeline, margin opportunities, cash requirements, customer risk, supplier risk and operational bottlenecks
• Design business-analysis and SWOT capabilities for the company, individual business lines and key customers
• Support budget-allocation and financing decisions with data-driven recommendations
• Ensure that Banner Copilot does not only display information, but proposes concrete next actions
• Engineering execution
• Collaborate with the lead engineer and developers
• Participate in implementing critical components and prototypes
• Conduct architecture and code reviews
• Establish technical standards and engineering practices
• Help decompose the architecture into deliverable phases and tasks
• Ensure implementation conforms to the approved architecture
• Current system modules
• ERP ingestion: MySQL adapter, RFQ input, daemon and manager
• Vendor selection: DuckDB matching, ranking, blacklist, database of 9,000+ suppliers
• Messaging: Gmail service account, email generation, dispatcher, deduplication, and cooldown
• Reply processing: Gmail auditor, PDF/XLSX/OCR parsing, Claude‑based extraction, confidence scoring
• Storage: SQLite WAL, sent registry, async audit logger
• API and interfaces: FastAPI, Tabulator.js, Sent Emails Dashboard, RFQ Workbench

Requirements

• 5+ years of backend development and architecture experience
• Strong Python: async, FastAPI, SQLAlchemy or similar tools
• Experience designing systems with LLM APIs, AI agents, and tool use
• Understanding of event‑driven architecture, queues, and asynchronous pipelines
• Experience designing integrations with ERP, CRM, or other enterprise systems
• Experience evolving an MVP or monolith into a scalable architecture
• Ability to judge where to apply AI versus conventional business logic
• Experience with production systems and real business transactions
• Ability to translate business problems into architecture and a coherent technical roadmap
• Experience designing decision engines, recommendation systems or scoring logic.
• Ability to work with incomplete, inconsistent and operational business data.
• Ability to communicate with business users, account managers and senior management.
• Ability to translate business goals into operating workflows, system architecture and phased implementation.
• Understanding of financial and operational decision factors such as margin, cash flow, financing needs and resource constraints.
• Working proficiency in English

Nice to have

• Experience in procurement, B2B sales automation, or supply chain
• Knowledge of aviation MRO and the aircraft components market
• Experience with the Claude API
• Prompt engineering and confidence‑scoring experience
• Knowledge of ILS, PartsBase, or similar aviation platforms
• Experience building human‑in‑the‑loop processes
• Experience developing pricing, recommendation, or decision engines
• Experience with web scraping and real‑time price retrieval
• Experience with account-management automation
• Experience with cash-flow forecasting, budget-allocation tools or financial decision support
• Experience building systems that generate management recommendations
• Experience with operational analytics and business-process redesign
• What we offer

What we offer

• Involvement in a strategically important AI product
• Opportunity to influence system architecture and long-term development
• Flexible schedule and remote work
• A culture that encourages experimentation, R&D, and implementation of new approaches
• Room for professional growth