Jeevan Babu V C

Experienced and detail-oriented professional with over 8 years of expertise in automation, functional testing, and system optimization. Skilled in leveraging tools such as Selenium, Playwright, and AWS to improve software quality, testing accuracy, and overall process efficiency. Adept at applying AI and ML techniques to strengthen workflows and deliver data-driven solutions. Recognized for strong problem-solving skills, adaptability in fast-changing environments, and a collaborative approach that drives continuous improvement and reliable outcomes.

• Recognized with Rising star award for ramping up the process in short span.
• Awarded as 'Einstein' for implementing maximum number of innovative ideas.
• Achieved Best performer award thrice in a row.

Agentic AI System for Autonomous Ticket Quality Monitoring and Risk Detection

• Built an AI agent that monitors resolved support tickets, spots quality risks and failure patterns on its own — cutting manual review work by 90%
• Added self-healing logic where the agent tracks its own failures, raises alerts after 10 consecutive issues, and stops processing after 50 unanalyzed tickets — no human needed
• Taught the agent to recognize ticket patterns like reopens, duplicates, and routing errors across 6 scenarios, so it picks the right analysis path automatically — filtering out ~18% of noise
• Made the agent smarter for high-priority tickets (Sev-1/2/2.5) by adding two specialized checks — one for routed tickets and one for resolved ones — covering RCA, MCM references, and routing quality
• Set up a config-driven prompt system so the agent adjusts its quality rules and SLA expectations per team — works across 5+ resolver groups without touching the code
• Wired up Slack alerts and scheduled reports so the agent proactively flags risks and patterns to 50+ engineers in real time

AI Agent Skill Validation, Defect Detection & Quality Engineering

• Performed full QA on a live AI agent handling seller barcode appeals — found 6 critical bugs that existing tests had completely missed
• Caught a routing bug where the agent was asking users to decide the workflow manually; fixed it to auto-detect from ticket data — cut handling time by ~30%
• Found a validation gap where a 4-point checklist was letting bad affiliation letters through; rewrote it as a proper 5-step pass/fail check — reduced false approvals by ~40%
• Spotted a missing duplicate-check causing the agent to re-approve already approved GTINs; added a live state lookup to skip them — saved ~25% of redundant processing
• Built and ran 100+ ground truth test cases using real production data, validated with both exact match and LLM-as-Judge scoring — hit 100% pass rate before go-live
• Pushed all fixes through a 10-step CI/CD quality gate — security scan, compliance checks, build validation — zero violations, clean release

Data Science and AI | Boston Institute of Analytics

Core Topics: Machine Learning, Natural Language Processing, Deep Learning.

Tools & Technologies: Python, Pandas, ML Libraries, NLP Frameworks, SQL

• Built predictive model to forecast shipping delivery times using historical e-commerce data
• Compared 4 ML algorithms (Logistic Regression, Random Forest, Decision Tree, SVM). Achieved 68% accuracy using optimized Support Vector Machine model
• Performed hyperparameter tuning with GridSearchCV, improving Random Forest to 68.4% accuracy

• Swimming
• Volleyball