Making an AI product more reliable before production

Product quality, functional testing and incident coordination for integrated AI workflows.

LLM QAAPISalesforceLinear

200+

Incidents tracked

Dev → Prod

Stages

Sandbox & API

Scope

The work

Challenge

Secure AI-product workflows where edge cases, integrations and LLM outputs could degrade the experience before production.

Approach

Turn test feedback into actionable tickets and create a common language across product, technical and operations teams.

Contribution

I tested workflows in the Sandbox, checked API and Salesforce integrations, evaluated LLM prompts and followed 200+ incidents in Linear through to resolution.

Impact and learning

Outcome

Continuous visibility on workflow quality and more usable feedback to prioritise fixes before deployments.

Key learning

For an AI product, quality is not limited to the model: it is also built through input data, integrations, test scenarios and clear feedback.

“This case study presents responsibilities, methods and non-confidential information only.”

Confidentiality note, Portfolio

My approach

Formalised test scenarios covering business workflows and edge cases.
Checked LLM outputs for consistency and usability.
Qualified, documented and tracked incidents in Linear.
Coordinated feedback across Sandbox, API, Salesforce and internal teams.

Context

This role focused on preparing and monitoring the quality of an AI product. Examples and metrics are presented only at a non-sensitive level: no customer data, internal screenshots or proprietary configuration is published.

Contribution

My role was to make anomalies understandable and actionable: identify the affected workflow, describe observed behaviour, specify the expected result and help teams track the fix through validation.

What this demonstrates

This experience reflects my interest in roles at the intersection of data, product and operations: keeping business reality visible while working with the technical constraints of an AI system.