Md Asaduzzaman

Solution‑focused professional with a strong background in designing, implementing, and managing complex IT solutions, now specializing in Generative AI and multi‑agent applications. Experienced in building agentic systems such as Employee Performance Agents and Order Fulfillment Agents to streamline store operations. Actively contributed to proofs of concept in audio transcription, diarization, and sentiment analysis using Azure OpenAI models. Successfully implemented MCP server integration with natural language query tools, enabling secure and validated communication with PostgreSQL. Adept at collaborating with stakeholders to deliver tailored architectures, skilled in cloud services, software development, and integration, ensuring optimal performance, scalability, and enterprise impact.

• Databricks Certified Generative AI Engineer Associate
• Databricks Certified Data Engineer Associate
• DP-700: Microsoft Certified Fabric Data Engineer Associate
• Databricks Certified Data Engineer Professional

• Engineered a Natural Language Query Tool enabling developers to interact with PostgreSQL databases conversationally — leveraging schema-aware context understanding to auto-generate, validate, and execute SQL queries in real time, eliminating the need for repetitive API development and significantly accelerating data access workflows.
• Delivered a POC for an AI-powered Conversation Intelligence Agent that transcribes, diarizes, and performs sentiment analysis on designer-customer interactions — autonomously extracting entity-level insights, evaluating designer performance, measuring customer satisfaction, and classifying conversation outcomes as Converted / Follow-up / Non-Converted.
• Built an AI-driven Employee Performance Agent that ingests live associate KPI data, applies intelligent recommendation logic to surface training and task management actions, routes manager alerts for approval-based execution, and converts audio feedback on new hires into structured KPI matrices for performance evaluation.
• Engineered an Order Fulfillment Agent that automated SLA-driven batch assignment for store associates, reducing pick-path distance by optimizing order grouping across SKU, category, and sub-category dimensions and mapping batches to in-store navigation routes for shortest-path picking
• Developed and managed ETL workflows for large datasets across multiple partner brands, ensuring high data accuracy and seamless integration into business processes.
• Performed Level 1 and Level 2 data transformations for intra-brand and cross-brand deduplication, significantly enhancing data consistency and creating golden records in Cosmos DB (No SQL DB)
• Collaborated with product teams, data scientists, and business stakeholders to analyze evolving requirements and design data engineering and API integration solutions aligned with business objectives.
• Developed and implemented a restriction on exploiting email claim journey for Gmail email IDs.
• Collaborated effectively with cross-functional teams, including QA, DevOps, and analytics teams, to ensure smooth project delivery and alignment across departments.
• Demonstrated strong analytical thinking, debugging, and problem-solving abilities to quickly identify issues and implement timely solutions, recognized with multiple appreciations for responsiveness and quality delivery.
• Prevented DDOS attacks on OTP-led journeys for international OTP, enabling runtime configuration to disable OTP generation for affected country codes, saving substantial costs.
• Revamped address module, creating a central repository of correct addresses, improving food delivery timeliness, and enhancing customer experience.
• Using partition key search in Cosmos DB cut response times from 3-digit to single-digit milliseconds, improving performance and reducing infrastructure costs for supporting 35M customers and 40K users/sec.
• The log sampling implementation reduced operational costs by 50%, lowering the average monthly cost from 3.3 million INR to 1.6 million INR per squad, with significant savings across 7-8 squads.
• Designed and implemented an AI-driven log analysis system by exporting Application Insights diagnostic data to Azure Blob Storage, generating schemas from dimensions logs, and storing log embeddings in a vector database. Developed an LLM-based analytics solution to enable intelligent querying, anomaly detection, and insight generation from application logs, improving observability and root-cause analysis efficiency.