Teradata
Software Engineer
- Architected an AI-driven ticket deflection system using LangChain, embeddings & MCP — automated 40% of support volume, saving 15+ engineering hours/week.
- Engineered a Text-to-SQL Engine with schema chunking & prompt optimization, reducing LLM token consumption by 35% while increasing accuracy on complex multi-table joins.
- Built Autonomous AI Agents integrated with Jira & Confluence APIs, deployed via Slack & Microsoft Teams for natural language enterprise operations.
- Re-engineered Backend Provisioning Workflows with task isolation & parallel execution — cloud setup time from 60s → 15s (75% faster).
- Optimized Approval Pipelines with delta-update processing — end-to-end latency from 40 minutes → <2 minutes (95% reduction).
- Productionized GitHub Actions image creation pipelines, ensuring 99.9% consistency across all development environments.