Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I’m Ishan Kumar, an AI Trainer and QA Lead with hands-on experience in multi-modal data annotation, LLM evaluation, and prompt optimization. I deliver guideline-driven, high-quality labels and clear evaluation reports across text, image, audio, and video data. My work focuses on improving model reliability by identifying hallucinations, bias, and safety risks, while refining annotation guidelines and mentoring annotators for better consistency and efficiency. I’m available for dataset creation and validation, LLM evaluation, prompt testing, quality audits, and building effective QC workflows.…I’m Ishan Kumar, an AI Trainer and QA Lead with hands-on experience in multi-modal data annotation, LLM evaluation, and prompt optimization. I deliver guideline-driven, high-quality labels and clear evaluation reports across text, image, audio, and video data. My work focuses on improving model reliability by identifying hallucinations, bias, and safety risks, while refining annotation guidelines and mentoring annotators for better consistency and efficiency. I’m available for dataset creation and validation, LLM evaluation, prompt testing, quality audits, and building effective QC workflows.

Ishan Kumar





I’m Ishan Kumar, an AI Trainer and QA Lead with hands-on experience in multi-modal data annotation, LLM evaluation, and prompt optimization. I deliver guideline-driven, high-quality labels and clear evaluation reports across text, image, audio, and video data. My work focuses on improving model reliability by identifying hallucinations, bias, and safety risks, while refining annotation guidelines and mentoring annotators for better consistency and efficiency. I’m available for dataset creation and validation, LLM evaluation, prompt testing, quality audits, and building effective QC workflows.…I’m Ishan Kumar, an AI Trainer and QA Lead with hands-on experience in multi-modal data annotation, LLM evaluation, and prompt optimization. I deliver guideline-driven, high-quality labels and clear evaluation reports across text, image, audio, and video data. My work focuses on improving model reliability by identifying hallucinations, bias, and safety risks, while refining annotation guidelines and mentoring annotators for better consistency and efficiency. I’m available for dataset creation and validation, LLM evaluation, prompt testing, quality audits, and building effective QC workflows.

Available to hire

I’m Ishan Kumar, an AI Trainer and QA Lead with hands-on experience in multi-modal data annotation, LLM evaluation, and prompt optimization. I deliver guideline-driven, high-quality labels and clear evaluation reports across text, image, audio, and video data. My work focuses on improving model reliability by identifying hallucinations, bias, and safety risks, while refining annotation guidelines and mentoring annotators for better consistency and efficiency. I’m available for dataset creation and validation, LLM evaluation, prompt testing, quality audits, and building effective QC workflows.

Skills

Experience Level

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Work Experience

QA Lead at Avis Budget Group

March 1, 2022 - Present

QA Lead responsible for AI-augmented testing, including Ollama-based local LLM execution to generate test cases and synthetic data; led a 7-member QA team in end-to-end functional and non-functional testing of microservices; designed and executed JMeter scripts for stress, spike, load balancing, scalability, fault/endurance testing; built an automation framework using Java, REST Assured, TestNG, Maven, Git and Extent Report for robust API testing; implemented observability with Grafana and Dynatrace for real-time diagnostics; provided stakeholder reports and led code refactoring and infrastructure optimizations for performance.

SDET at Velir Inc

March 1, 2015 - October 1, 2018

API/Web Automation: Sitecore RESTful API to create/delete/update website items; API/Web testing using Postman, REST Assured, and Selenium WebDriver. Performance Testing: Developed load testing scripts using Apache JMeter and BlazeMeter for performance validation. Observability & Monitoring: Analyzed system behavior under load using New Relic for real-time metrics, identifying bottlenecks. Scalability Testing: Validated horizontal/vertical scaling capabilities under peak traffic to ensure seamless user experience.

Intern at BlackBerry

January 1, 2014 - June 1, 2014

Mobile Automation: Developed keyword-driven automated tests for Hub mobile application using Robot Framework. Test Case Design: Created test cases for the BlackBerry Intelligent Assistant feature by analyzing User Experience documents. Data Validation: Created and executed SQL queries for SQLite databases for root cause analysis and data validations.

SDET at Citibank (via Accenture)

April 1, 2011 - November 1, 2012

Web Automation: Developed automated test scripts for web-based loan applications using UFT (VBScript). Test Case Design: Created test cases in HP ALM for CitiBank applications by analyzing functional requirements. Defect Management: Discovered, investigated, reported, and tracked software issues throughout the SDLC.