Job Description

This position will lead the technical design, deployment, and operational management of QAD's Finance AI agents. The role bridges finance operations, data engineering, and AI systems to build secure, high-performing, and auditable agent-assisted workflows that optimize finance processes.

Core Responsibilities:

1. Data Foundation & Semantic Layer

- Define and own the context tables (business glossary, entity definitions, metric hierarchies) that ground AI agents in QAD's financial reality.

- Build and maintain the semantic layer — the translation between raw Google Cloud BigQuery datasets and the financial concepts agents reason over

- Partner with the data engineering team and other functions on dbt model governance and Fivetran pipeline integrity

2. Agent Architecture & Process Design

- Determine the build methodology for process agents — frameworks, orchestration patterns, tool use conventions, handoff protocols between agents

- Sequence the agent build roadmap.

- Design the human-in-the-loop thresholds — when an agent escalates vs. executes autonomously

- Own the technical architecture decisions: LLM selection, retrieval strategy, memory design, context window management

3. Agent Operations & Lifecycle Management

- Run the agent registry — versioning, deployment, deprecation, and change control for all live agents

- Manage agent performance over time: drift detection, accuracy degradation, prompt updates as business rules change

- Own the incident response protocol when an agent produces an incorrect output or takes an unintended action

- Coordinate agent updates when underlying data schemas, business logic, or finance policies change

4. Visibility & Monitoring Platform





- Build or procure a Finance AI Control Center — a dashboard giving real-time visibility into which agents are running, what they've done, error rates, and escalation queues

- Define and track agent KPIs: task completion rate, exception rate, cycle time vs. manual baseline, cost per transaction

- Create an audit log for every agent action — who initiated it, what data was touched, what output was produced

- Report agent health and ROI to management on a regular cadence

5. AI Risk Management & Internal Controls

- Design and own the internal control framework for AI in Finance

- Define approval hierarchies: what an agent can do autonomously vs. what requires human sign-off (e.g., any journal entry over $X requires controller review)

- Partner with infosec on data access controls

6. Finance Organization Training & Change Management

- Design and deliver the Finance AI literacy curriculum — from foundational concepts to role-specific agent interaction training

- Build the change management playbook for the transition: how teams move from manual processes to agent-assisted workflows

- Create and maintain agent user guides — what each agent does, its limitations, how to override it, and how to report issues

- Work with Finance leaders to embed AI fluency into their team cultures

7. Process Reengineering

- Lead the Finance process rewrite initiative

- Identify which steps in each process are agent-executable vs. requiring judgment, and draw those boundaries explicitly

- Ensure redesigned processes still satisfy internal control requirements and auditability standards

- Feed process redesign outputs back into agent specifications and the semantic layer

📌 Lead AI Specialist (Finance Operations) (Mumbai)
🏢 QAD
📍 Mumbai