Business Analyst
Location & Logistics
- Location: Los Angeles area
- Onsite cadence: 4 days per week onsite
- Position type: Contract-to-hire
- Experience target: Open — junior to mid-level acceptable for the right AI-native profile
About the Role
The hiring team is consolidating analyst, procurement-analyst, and IT functions into a new Business Intelligence group, and this seat is the AI-native half of the foundation. The data engineer next to you owns the pipes; you own the thinking on top of them. The hiring manager said it directly: the era of analyst-as-report-builder is ending — what they need is someone who treats Claude and ChatGPT as a daily tool, builds custom MCP connectors to plug LLMs into the ERP and the CRM and the cost models, and turns the org's stack of fragile Excel files into automated, queryable workflows.
You'll spend four days a week at a plant in the LA area, sitting next to operations and procurement, learning the business by walking the floor and reading the spreadsheets nobody has ever documented. The other half of the job is technical: building decision-oriented analytics, designing the company's first proper set of business and department-level KPIs, and replacing whole categories of manual analyst work with agentic and AI-assisted automation.
This is not a "I made a chart in Power BI" analyst seat. It's the seat for an analyst who's watched the field shift under their feet, leaned into it, and now operates at a level most five-year veterans don't.
Responsibilities
- Build custom MCP-style connectors between Claude / OpenAI and operational systems (ERPs, ADP, costing models, spreadsheets, CRM)
- Replace fragile Excel costing models with automated, durable workflows
- Design and roll out a full set of business KPIs and department-level KPIs across operations, procurement, and sales
- Stand up decision-oriented analytics — not "here's the data," but "here's the decision and why"
- Partner with the data engineer to close data quality gaps and turn untrustworthy reports into trustworthy ones
- Learn the operational business deeply enough to identify where AI-assisted automation will move actual outcomes
Qualifications
- A working pattern of using Claude, ChatGPT, or similar tools as a daily problem-solving instrument — not curiosity, fluency
- Hands-on experience building or wiring up custom connectors, agents, or LLM workflows (MCP, function calling, Zapier-AI, custom Python, etc.)
- Strong Excel modeling — comfortable inheriting messy spreadsheets and rebuilding them properly
- Comfort being onsite at a manufacturing plant 4 days a week in the LA area
- Self-starter wired for plant-floor, operations-adjacent work — energized by it, not endured
Preferred Skills
- Food, CPG, or manufacturing industry exposure
- ERP data fluency (Microsoft Dynamics, NetSuite, SAP, or similar)
- Light Python or Node scripting
- Prior experience replacing analyst headcount with automation
Tech Stack
Stack was rebuilt last year and is intentionally simple. Expect: Claude / OpenAI APIs, custom MCP connectors, a SQL-based warehouse owned by the data engineer next to you, BI/dashboarding tool, two ERPs, ADP, and a stack of Excel models on their way out.
Closing
This is the seat for the analyst who's already figured out that the future of the job is half builder, half operator. Four days a week on a real plant floor, AI-native tools in your hands, and a hiring manager who actively wants you to automate yourself out of the boring parts.
