Chandra T

Develop end-to-end Generative AI workflows that transformed large volumes of structured and unstructured enterprise data into actionable intelligence for decision-making applications. Built retrieval-augmented pipelines that combined semantic search, dense vector embeddings, and large language models to enable accurate question answering, summarization, and insight generation. Developed and fine-tuned NLP models to extract entities, relationships, and signals from free-text documents, enabling richer analytics and downstream intelligence. Implemented real-time inference flows that produced concise summaries, risk indicators, and recommended actions, while applying structured prompting, confidence scoring, and validation checks to ensure reliable and explainable outputs. Established continuous monitoring for response quality, latency, retrieval relevance, and model drift, and maintained freshness through ongoing data ingestion, embedding refresh, and model updates. Engineered scalable data pipelines to ingest, cleanse, normalize, and index high-volume records and documents, creating reusable feature datasets that blended behavioral, transactional, and contextual signals. Enforced strict data governance, privacy controls, and access policies before exposing data to analytics and AI systems. Collaborated closely with business users, subject-matter experts, and governance teams to validate outputs against real-world scenarios, refine workflows, and align solutions with operational expectations. Contributed to reusable patterns, standards, and best practices for responsible AI delivery, enabling rapid onboarding of new use cases and establishing a sustainable foundation for long-term AI adoption across the organization.

Designed and implemented an end-to-end data and analytics and AI automation workflow that transformed large volumes of structured and unstructured enterprise data with high-volume transactional records into trusted, analytics-ready datasets and real-time risk insights. Built ingestion flows capable of handling both streaming and batch data, followed by robust validation, standardization, and enrichment steps to ensure data quality, consistency, and auditability before downstream use. Developed unified data models representing entities, events, and relationships, enabling consistent reporting and cross-functional analysis. Created curated analytical layers and feature datasets that supported monitoring, scoring, and decisioning workflows. Engineered feature-generation processes that captured behavioral patterns, trends, and anomalies, feeding predictive models and rule-based logic used in operational systems. Implemented automated scoring pipelines that applied trained models to incoming data and persisted results for analytics, monitoring, and compliance review. Established continuous checks for data quality, feature stability, and model performance to detect shifts in behavior and maintain reliability over time. Optimized processing and storage patterns to support low-latency querying, dashboards, and API-driven consumption. Enforced strong data governance through access controls, masking, and segregation policies, ensuring sensitive data was protected while remaining usable for analytics. Enabled transparency through data cataloging, lineage, and ownership tracking. Delivered executive- and analyst-ready dashboards and reports, and coordinated the full lifecycle using automated orchestration, version control, and CI/CD practices while collaborating closely with cross-functional teams in iterative delivery cycles.

Designed and implemented a centralized data workflow that unified datasets from multiple departments into a single, governed analytics foundation. Built controlled ingestion processes to extract data from legacy systems and stream operational events, ensuring secure, auditable, and reliable data movement. Developed large-scale transformation pipelines to cleanse, standardize, and integrate diverse datasets with varying structures and quality levels, producing consistent and analytics-ready outputs. Established strong governance and transparency by maintaining metadata, lineage, ownership, and documentation across the data lifecycle, enabling audit readiness and long-term maintainability. Engineered both batch and near–real-time processing paths to support operational updates alongside historical analysis. Implemented event-processing pipelines to filter, enrich, and route streaming data before aggregation, ensuring low-latency and high-quality delivery to downstream workflows. Created structured analytical datasets optimized for reporting, compliance reviews, and policy-driven analysis, while applying performance optimizations to handle growing data volumes efficiently. Embedded reconciliation, validation, and regression checks to ensure financial accuracy and cross-system consistency. Enforced fine-grained security and access controls to protect sensitive records and ensure departmental data segregation. Collaborated closely with multiple stakeholder groups to standardize definitions, metrics, and reporting logic across organizations. Orchestrated end-to-end workflows with automated scheduling, dependency management, and controlled deployments. Enabled self-service analytics and dashboards for analysts, auditors, and leadership, supporting informed decision-making and regulatory confidence through reliable, transparent, and well-governed data pipelines.

Designed and delivered a scalable data workflow to ingest, process, and analyze extremely high-volume operational and transactional records generated by network and billing systems. Built ingestion pipelines capable of handling both scheduled batch loads and continuous event streams, ensuring timely availability of raw data for downstream processing. Implemented orchestration workflows to coordinate ingestion, distributed processing, validation, and publishing steps with clear dependencies and recovery mechanisms. Developed large-scale transformation pipelines to decode proprietary data formats and enrich raw events with subscriber, device, and location attributes. Structured datasets using efficient storage layouts and partitioning strategies to support high-performance querying and cost-efficient storage. Embedded data-quality, reconciliation, and anomaly-detection checks to validate consistency between operational events and financial records, enabling early issue detection. Created aggregation pipelines to compute usage, quality, and revenue-related metrics used for operational monitoring, billing validation, and performance analysis. Optimized distributed processing through tuning strategies that reduced execution time and improved reliability under heavy data volumes. Exposed curated datasets through SQL-accessible interfaces to support ad-hoc analysis and standardized reporting. Enabled secure, self-service analytics for business and operations teams while enforcing strict access controls and data protection policies. Delivered trusted datasets supporting fraud analysis, churn insights, and regulatory reporting. Collaborated closely with engineering and operations stakeholders to validate metric definitions, align outputs with service-level expectations, and maintain detailed production documentation and runbooks for long-term support and operational stability.

Designed and implemented a foundational data workflow to centralize operational, usage, and training-related datasets from multiple on-premise sources into a single analytics-ready environment. Built reliable batch ingestion processes to land raw logs, activity records, and certification data into structured landing zones, followed by transformation steps to cleanse, standardize, and enrich semi-structured and structured datasets. Developed reusable ETL scripts to curate raw data into relational models that supported reporting on platform utilization, resource consumption, and operational performance. Implemented scheduled processing jobs to generate daily aggregations and key metrics, ensuring consistent and timely data availability for analytics users. Optimized data models and query performance through indexing and efficient SQL design to support interactive reporting workloads. Embedded basic data-quality and reconciliation checks to validate completeness and consistency between source systems and curated datasets. Enabled secure, role-based access to analytical data, ensuring appropriate segregation for reporting and analysis. Supported ad-hoc analysis and dashboarding by exposing well-defined reporting views and query-ready tables. Collaborated closely with infrastructure and operations teams to validate metrics, resolve data issues, and lay the groundwork for early monitoring and analytics initiatives through a reliable, maintainable data foundation.