AI Engineer with 1+ years in Generative AI & Agentic AI Development
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Generative AI Engineer and Agentic AI Developer with 1+ year of experience designing and deploying production-grade multi-agent AI systems for enterprise clients across banking and aviation domains. Deep expertise in LLM orchestration, Retrieval-Augmented Generation (RAG), prompt engineering, and agentic workflows using AWS Bedrock, LangGraph, LangChain, and Anthropic Claude. Proven track record of delivering measurable outcomes: 90% AHT reduction, >98% LLM extraction accuracy, and 65% FTE effort reduction at scale. Experienced in responsible AI, AI safety, guardrails, LLM evaluation, and compliance-ready deployments.
D.Y. Patil College of Engineering
Bachelor of Engineering · Artificial Intelligence and Data Science
August 1, 2021 – May 1, 2025
WNS-Vuram
Associate Technical Consultant - AI Engineer (Generative AI)
June 1, 2025 – Present
Pune, Maharashtra, India
CardOpsAI Agentic AI Multi-Agent Back-Office Automation
June 24, 2026 – Present
Architected a production-grade agentic AI system using AWS Bedrock multi-agent orchestration (Supervisor-Agent pattern) comprising Split Request, Extraction, Verification, Action, and Evaluator agents automating high-volume card operations (limit increase, address change, card block, replacement) end-to-end for a banking enterprise client. Engineered multimodal LLM extraction pipeline with >98% accuracy on unstructured emails and documents using Anthropic Claude 4.5/3.7, integrated via secure REST to core banking and card-processor APIs with full business-rule validation. Implemented Human-in-the-Loop (HITL) via Appian with an Evaluator agent enforcing compliance checks and auto-initiating re-runs on policy violations delivering SLA-compliant, fully auditable responses with end-to-end traceability. Achieved 90% AHT reduction (30 min → <3 min), 30-50% TAT improvement, 40-60% self-service deflection, and ~65% FTE effort reduction through queue-based serverless execution with dead-letter handling.
Agent CentralAI AI Monitoring, Evaluation & Governance Platform
June 24, 2026 – Present
Built a near real-time AI observability and governance platform for production AWS Bedrock agents across the full lifecycle (registration, testing, deployment, operations) — aggregating 15+ LLM evaluation metrics: Hallucination Risk, Prompt Injection, PII Detection, Context Adherence, Toxicity, Tool Selection Quality, and Completeness. Engineered automated responsible AI scanning for injection/leakage risks and guardrail optimization across action group Lambdas and knowledge bases, with actionable remediation and inclusiveness/privacy/fairness scoring. Delivered A/B testing for safe agent version comparison pre-rollout, cost analytics with anomaly alerts, CloudFormation IaC deployment, and a unified compliance dashboard — forming a single system of record for 365-degree agent governance.
Flight Disruption Management (FDM) AI System
June 24, 2026 – Present
Designed a production multi-agent conversational AI for WestJet Airlines using LangGraph and Claude 3.5 (Supervisor-Agent pattern) — automating flight cancellation and rebooking workflows with NLP-based intent detection and PNR extraction across multi-turn dialogues. Implemented RAG using FAISS vector store and Azure OpenAI embeddings to validate requests against airline SOPs in real time, with dynamic penalty calculation and complex edge-case handling (departed flights, multi-passenger bookings).
VybeStudio AI-Powered Mockup Generation System
June 24, 2026 – Present
Architected a 6-node LangGraph agentic workflow (input processing, story analysis, architecture planning, page generation, screenshot capture, validation) for automated UI mockup generation from textual requirements — reducing design time by 65% and delivering 120+ mockups in a distributed execution environment. Implemented computer vision-based consistency validation with automated screenshot comparison achieving 95% inconsistency detection; designed shared global state caching for cross-page contextual awareness, improving output consistency by 40%.
AWS Certified AI Practitioner
AWS
June 1, 2026 – Present
Cultural Fit Analysis
The candidate's project portfolio showcases diverse applications of AI in banking, aviation, and UI generation, indicating adaptability and a broad interest in solving real-world problems across different domains. Their experience with multi-agent systems and collaborative tools (e.g., Appian for HITL) suggests a team-oriented approach. The focus on responsible AI and governance aligns with a culture that values ethical development and robust operational practices. The candidate's proactive approach to certifications and continuous learning also indicates a strong cultural fit for an innovative and growth-oriented environment.
Soft Skills & Operational Fit
The candidate demonstrates strong problem-solving skills through complex multi-agent system designs and a clear understanding of operational metrics (AHT, TAT, FTE reduction). Their experience with Human-in-the-Loop (HITL) systems and compliance checks indicates an appreciation for robust, auditable, and production-ready solutions. The emphasis on responsible AI and governance suggests a proactive approach to ethical and safe AI deployment, which aligns well with modern operational best practices.