Founding Applied AI Engineer

DOSS.COM; San Francisco (CA); On-site; USD 180,000 - 280,000; Full time

Job summary

Doss is building an AI-powered Operations Cloud for real-world product businesses in San Francisco. You will own applied AI across Doss’ platform—Dossbot, schema/workflow generation, and agentic automation that work with live procurement, inventory, and finance data.

As a senior engineer, you’ll design end-to-end AI architecture, ensure safety with guardrails, and mentor peers to raise the field’s bar for enterprise AI. This role sits at the core of transforming back-office processes.

Qualifications

• Shipped LLM-powered features or agents to production.
• Hands-on work with foundation model APIs, RAG, tool use, prompt and context engineering.
• Built eval harnesses and guardrails for non‑deterministic systems running against business‑critical data.
• Strong backend fundamentals: designing systems, Postgres, and cloud services for reliable, multi‑tenant web applications.
• Judgment for when AI is the right tool and when it isn't.

Responsibilities

• Own the applied AI architecture end-to-end: model selection, orchestration, RAG over the unified data model, evals, guardrails, and cost/latency budgets in production.
• Build Dossbot into the primary interface for operations: natural-language search, analysis, and automation grounded in live supply-chain, finance, and ops data.
• Engineer AI-assisted tooling that uses foundation models to propose, validate, and apply customer schemas and workflow definitions on live production tenants.
• Design agentic workflows that act inside enterprise guardrails: permission-aware, auditable, tenant-isolated, and safe to run against a customer's system of record.
• Stand up the eval and observability infrastructure to measure quality, catch regressions, and proactively detect anomalous or risky emergent behavior.
• Partner with product, design, and our Forward Deployed team to deliver on their highest-leverage AI use cases.
• Mentor and shape engineering culture as the team's center of gravity for applied AI, raising the bar on how we build, evaluate, and ship LLM-powered systems.

Skills

LLM features in production

Foundation models

RAG pipelines

Prompt engineering

Guardrails & safety

Backend design

Tools

Postgres

Cloud services

Job description

About Doss

DOSS is building an Operations Cloud for the real world.

A modern, AI-native platform for physical product businesses to manage the flow of goods, dollars, and data in real time, across:

• Procurement

• Inventory

• Orders

• Fulfillment

• Finance

Built to replace spreadsheet chaos and rigid, consultant-heavy ERPs. Fast time-to-value, adaptable as the business evolves.

Through our Adaptive Resource Platform (ARP) and unified operational data model, teams deploy quickly, automate workflows, and make changes without months of re-implementation.

We recently raised a $55M Series B, co-led by Madrona and Premji Invest. Participation from Intuit Ventures, Theory Ventures, General Catalyst, Contrary Capital, and Pathlight VC.

DOSS is trusted by fast-growing operators to run critical operations with speed, control, and confidence.

About the role:

Every ERP claims to be "AI-powered." Almost none of them let an operator ask a question in plain language and get an answer grounded in their live procurement, inventory, and finance data, let alone let an agent safely take action on it. That's the goal of this job: you'll own applied AI across Doss: Dossbot (query, analyze, and automate via chat), AI-assisted schema and workflow generation, and the agentic automations that turn back-office busywork into background processes. You'll be the technical anchor for how foundation models meet a transactional system of record.

What you’ll do:

• Own the applied AI architecture end-to-end: model selection, orchestration, RAG over the unified operational data model, evals, guardrails, and cost/latency budgets in production.

• Build Dossbot into the primary interface for operations: natural-language search, analysis, and automation grounded in live supply-chain, finance, and ops data.

• Engineer AI-assisted tooling that uses foundation models to propose, validate, and apply customer schemas and workflow definitions on live production tenants.

• Design agentic workflows that act inside enterprise guardrails: permission-aware, auditable, tenant-isolated, and safe to run against a customer's system of record.

• Stand up the eval and observability infrastructure to measure quality, catch regressions, and proactively detect anomalous or risky emergent behavior.

• Partner with product, design, and our Forward Deployed team to deliver on their highest-leverage AI use cases.

• Mentor and shape engineering culture as the team's center of gravity for applied AI, raising the bar on how we build, evaluate, and ship LLM-powered systems.

What we’re looking for:

Experience

• Shipped LLM-powered features or agents to production and operated them at scale, not just prototypes or demos.

• Deep hands‑on work with foundation model APIs, RAG pipelines, tool use/function calling, prompt and context engineering, and fine‑tuning where it earns its keep.

• Built eval harnesses and guardrails for non‑deterministic systems running against business‑critical data.

• Strong backend fundamentals: Designing systems, Postgres, and cloud services for reliable, multi‑tenant web applications.

• Going from 0 to 1 in ambiguous settings, and judgment for when AI is the right tool and when it isn't.

Qualities

• Team player: you bring positivity, openness, and curiosity to the team every day.

• Growth mindset: everything is an opportunity to learn and improve yourself, the team, and the company.

• Craftsmanship: you care about making something of quality.

• Customer First: you care immensely about the customer experience journey and want to create a product full of delight.

• Pragmatic: you make the right decision based on circumstance, not theory.

• Low Ego: you work best in collaborative environments where everyone supports the team.