Gunavant Desale

QA and process documentation specialist with direct experience turning ambiguous, gap-ridden specifications into clear, consistently applicable standards.

At Kotak Life

Insurance, I identify recurring process gaps and rewrite them into official team checklists that others follow without escalation — the same discipline required to close the gap between program requirements and rater-ready instructions. Combined with active experience following structured AI evaluation rubrics (Alignerr AI, Apple Maps/CrowdGen), I understand both sides of the rubric relationship: writing clear guidance, and applying it precisely as a rater myself.

AI Linguistic Specialist - Apple maps (via appen)

(2026-06)

• Annotate and evaluate Hindi-language AI speech output by Siri for Apple Maps navigation — applying native-language judgment to assess accuracy, consistency, and relevance against detailed project guidelines.
• Document pronunciation errors and edge cases with structured written feedback, identifying failure patterns to improve annotation standards and support model retraining.
• Maintain consistent annotation quality across high-volume review sessions, adapting to evolving guidelines as project requirements are refined.

AI Data Annotation Reviewer - Alignerr AI

(2025-08 - 2026-05)

• Apply structured evaluation rubrics to rate AI-generated video content — consistent, guideline-driven rating across high-volume review cycles, with structured escalation when guidelines don't cover an edge case.
• Annotate and review AI-generated video content against structured guidelines — labelling action accuracy, object interaction, and timing precision across high-volume daily tasks.
• Apply critical thinking to ensure annotation labels are accurate, consistent, and contextually relevant; document ambiguous cases and escalate for guideline clarification.
• Provide structured feedback and process improvement suggestions that refine annotation standards and reduce future ambiguity for the annotation team.

Associate Lead - QA & Delivery Operations - KLI

(2024-10 - 2025-08)

• Identified 15+ recurring process gaps per quarter where existing documentation was ambiguous or contradictory — root-caused each one and rewrote it into the team's official QA checklist, eliminating the need for repeated escalation.
• Before/after example: sprint readiness criteria were previously interpreted inconsistently across reviewers, causing repeated escalations; rewrote the criteria into an unambiguous, example-anchored checklist that reviewers now apply independently.
• Own QA documentation across 4+ concurrent enterprise insurance product releases — direct exposure to compliance-sensitive, domain-specific specification writing in a regulated financial services environment.

QA Intern - QK

(2024-02 - 2024-09)

• Executed 200+ structured test cases per sprint with zero-ambiguity documentation — developing the meticulous attention to detail and consistent labelling discipline central to data annotation work.

B.tech - Computer Engineering - RCPIT, Shirpur, Dhule (2024-07)