Mathanagopal Sadagopan

About

AI/backend engineer who independently built and operated an agentic telecom analytics system end-to-end, using LangGraph and Claude to turn natural language into safe SQL in a regulated environment. He combines startup-speed execution with compliance-minded rigor, citing 95%+ NL-to-SQL accuracy, a 30-minute-to-2-minute workflow improvement, and zero-findings support across three regulatory audit cycles.

Key Strengths

• End-to-end ownership of production backend and AI systems
• Built agentic NL-to-SQL system for telecom operations
• Reduced incident diagnosis workflow from 30 minutes to under 2 minutes
• Achieved 95%+ NL-to-SQL query success across 4 enterprise operational domains
• Strong reliability-first engineering in regulated environments
• Implemented schema-first validation and structured output guardrails
• Independent production deployment and operations ownership
• Resolved 6 post-launch production bugs with zero downtime
• Balances reliability, latency, and developer velocity pragmatically
• Ships quickly in 6-week MVP/iteration cycles
• Strong observability and incident-response mindset
• Used inter-model benchmarking to improve factual accuracy and safety
• Reduced escalation rate by 25% on upstream data mismatch issue
• Reduced investigation time by 40% through SQL anomaly detection
• Supported 3 consecutive regulatory audit cycles with zero findings
• Effective at defining scope in ambiguous early-stage environments

Experience

Education

Skills

Languages

Certifications