Senior AI/ML Engineer - LangChain/GenAI Specialist

Jhansi Priya

**********@*****.*** +1-210-***-****

Professional Summary

Senior AI/ML Engineer with 10 years of experience in AI solution development, LangChain/LangGraph frameworks, and Python-based ML model engineering.

Skilled at deploying scalable AI/ML solutions leveraging LangChain, LangGraph, and cloud-native pipelines across finance, healthcare, and automotive domains.

Strong foundation in SQL, data modeling, ETL design, lineage documentation, field level mapping, and data-quality engineering supporting enterprise analytics and AI workloads.

Expertise in building LLM/GenAI applications, RAG pipelines, semantic search systems, and vector-based retrieval workflows using modern AI frameworks.

Skilled in developing and deploying ML pipelines, including forecasting, classification, clustering, anomaly detection, NLP, and responsible AI governance.

Advanced experience across Azure, AWS, and GCP in ML deployment, data pipelines, orchestration, MLOps automation, and cloud native model lifecycle management.

Proven ability to translate business requirements into scalable data architectures, ML models, lineage mapping, and AI-driven insights with strong documentation and compliance alignment.

Recognized for bridging data engineering + machine learning + GenAI, delivering production-ready AI systems with emphasis on accuracy, reliability, and operational excellence.

Adept at architecting integrated AI ecosystems combining ingestion pipelines, feature engineering, LLM driven workflows, and automated model monitoring.

Strong background in semantic search, vector indexing, embedding based retrieval, and relevance ranking optimization for enterprise information access.

Experienced in implementing LLM safety frameworks, guardrails, data-masking controls, quality scoring, and drift detection for regulated AI environments.

Skilled in designing cross-cloud ML/GenAI architectures and optimizing workloads across Azure, AWS, and GCP.

Strong communicator capable of simplifying complex ML/LLM concepts, model performance, and data-quality insights for business and executive stakeholders.

Skilled in transforming legacy data systems into modern cloud-native ML platforms through scalable ingestion, feature store design, and container-based deployments.

Hands-on expertise integrating multimodal AI workflows including text, tabular, audio, and document-based LLM pipelines.

Effective collaborator working with engineering, compliance, security, and product teams to deliver secure, governed, and scalable AI/ML solutions.

Adept at optimizing cloud compute costs for GenAI and ML workloads through caching, distributed processing, GPU utilization, and pipeline tuning.

Proven success implementing end-to-end monitoring frameworks covering LLM quality, feature drift, data freshness, and inference reliability.

Strong experience evaluating vector similarity metrics, embedding models, and retrieval scoring techniques to improve RAG accuracy.

Ability to collaborate with product and engineering teams to define GenAI use cases, design prototypes, and translate them into production-grade systems.

Skilled in redesigning ETL/ELT workflows and ML pipelines to meet scalability, compliance, and high-availability requirements.

Experienced in validating AI systems for regulatory compliance across banking, healthcare, and government domains.

Core Skills:

Category

Technologies / Tools / Frameworks

Programming & Frameworks

Python, SQL, PySpark, R, TensorFlow, PyTorch, Scikit-learn, Keras, Hugging Face Transformers

Generative AI & LLMs

GPT Models, LangChain, LangGraph, LlamaIndex, OpenAI APIs, RAG Pipelines, Embeddings, Prompt Engineering, Vector Databases (FAISS, Pinecone, Chroma)

MLOps & Automation

Azure ML, AWS SageMaker, Google Vertex AI, MLflow, Kubeflow, Docker, Kubernetes, Microservices Architecture (MCP), Model Registry, CI/CD Pipelines, Monitoring & Drift Detection

Data Engineering & Storage

Apache Spark, Delta Lake, Databricks, Airflow, Kafka, BigQuery, Snowflake, Data Lakes, Distributed Processing, ETL/ELT Workflows

Cloud & Infrastructure

Microsoft Azure, AWS, Google Cloud Platform, Serverless Computing, Container Orchestration, Infrastructure Automation

AI Governance & Explainability

SHAP, LIME, Fairlearn, Responsible AI Controls, Bias Detection, Model Interpretability Dashboards, Data Lineage & Quality Validation

Natural Language Processing

Transformers, BERT, GPT, SpaCy, NLTK, Text Classification, Summarization, Semantic Search, NER, Sentiment Analysis

Computer Vision

CNNs, OpenCV, Image Classification, Object Detection, Segmentation, Vision Transformers (ViT)

Data Science & Analytics

Statistical Modeling, Predictive Analytics, Feature Engineering, Time Series Forecasting, A/B Testing, Experimentation, Business Insights

Collaboration & Dev Practices

Git, Jira, Confluence, Agile/Scrum, Documentation, Cross-Functional Collaboration, Technical Leadership

Professional Experience

Bank Of America Senior GenAI / ML Engineer August 2024 – Present

Designed and deployed enterprise-grade GenAI architectures integrating RAG, vector search, embeddings, and LLM-driven automation across fraud, credit, and compliance functions.

Led end-to-end development of LLM assistants that automated fraud case reviews, credit memos, risk summaries, and compliance documentation.

Developed and deployed LangGraph based conversational workflows and A2A integrations connecting LLM assistants with enterprise microservices.

Established GenAI governance standards including prompt controls, data-masking layers, audit logging, and safe-usage guardrails.

Architected AI/ML workflows integrating Azure based data pipelines with GCP Vertex AI model deployment for scalable GenAI applications.

Designed Dialogflow CX + Vertex AI based chat virtual agents supporting intent detection, slot filling, and contextual conversations

Built CCAI serverless architectures using Pub/Sub, Cloud Functions, and Cloud Run for real-time chatbot orchestration

Partnered with risk, compliance, and enterprise architecture to operationalize governed GenAI workflows aligned to regulatory expectations.

Built predictive ML models for fraud detection and creditworthiness using classical ML algorithms and PySpark pipelines.

Engineered highly scalable ETL workflows in Azure Synapse to unify multi-source financial datasets for model training and analytics.

Designed semantic search pipelines involving embedding optimization, chunking, hybrid search, and ranking evaluation.

Developed LLM performance dashboards analyzing precision/recall, retrieval quality, hallucination trends, and latency metrics.

Optimized LLM prompts and system templates tailored to financial-use-case constraints and accuracy requirements.

Implemented CI/CD pipelines with GitHub Actions for ML & GenAI workloads including automated validation gates and rollback conditions.

Built automated drift detection frameworks monitoring data quality, embedding drift, and model performance degradation.

Developed lineage maps, schema definitions, input-output field dictionaries, and audit-ready documentation for all ML assets.

Introduced monitoring hooks for LLM request success rate, token usage, cost optimization, and time-to-response metrics.

Worked with data engineering teams to improve upstream ingestion quality, schema consistency, and metadata completeness.

Reviewed ML/GenAI architecture, code, and deployment plans across teams to ensure best practices and platform consistency.

Provided technical guidance, onboarding, and mentorship to junior engineers and cross-functional analytics teams.

Hyundai Motor Group AI Automation Engineer November 2022 – July 2024

Built automated data pipelines ingesting EV and connected-car telemetry (battery stats, sensor logs, navigation traces, charging data) using AWS Glue, Lambda, and Kinesis.

Developed PySpark workflows to clean, enrich, and process high-volume vehicle time-series data for predictive analytics.

Built microservices-driven (MCP) AI components enabling real-time integration between telemetry analytics and predictive maintenance services.

Designed ML models automating early detection of EV battery degradation, component failure risk, and abnormal driving patterns.

Built anomaly-detection systems automatically flagging abnormal temperature spikes, charging irregularities, and sensor malfunctions.

Developed LLM-based service-automation workflows for technician support, repair-manual search, diagnostic guidance, and troubleshooting steps.

Implemented RAG pipelines retrieving service manuals, diagnostic codes, warranty rules, and EV technical bulletins through embeddings + vector DB.

Automated customer-support workflows using an LLM-powered assistant providing responses about maintenance schedules, range issues, and EV performance.

Created PHI/PII masking layers, prompt-sanitization logic, and LLM guardrails ensuring safe and compliant automation.

Built SQL-driven logic for automated vehicle segmentation, service-risk scoring, and usage-pattern analysis.

Automated feature pipelines integrating telematics, warranty claims, dealership repair logs, and charging-station data.

Designed dashboards enabling automated monitoring of EV fleet health, charging behavior, and predictive maintenance alerts.

Built CI/CD workflows for automation scripts, ML models, and LLM pipelines using Jenkins and GitHub Actions.

Developed drift-detection triggers automating retraining based on new sensor behavior, environmental conditions, or updated firmware.

Built reusable automation libraries for ingestion, validation, monitoring, and EV-signal computations.

Automated telemetry-quality checks detecting missing signals, inconsistent frequencies, and incorrect sensor readings.

Collaborated with EV engineers and digital service teams to operationalize automated insights into production platforms.

Authored documentation for automated workflows, LLM pipelines, lineage maps, and telemetry processing logic.

Led automation-focused cross-team initiatives improving efficiency across vehicle analytics, engineering, and service operations.

Molina Healthcare ML Engineer / Data Scientist September 2021 – October 2022

Developed ML pipelines supporting disease-risk prediction, care-gap detection, preventive care recommendations, and patient stratification.

Built LLM/NLP pipelines for clinical note summarization, physician documentation support, and automated care-plan generation.

Designed PHI-safe LLM workflows ensuring redaction, masking, and HIPAA-compliant prompt input sanitization.

Integrated real-time ML scoring into provider dashboards improving clinical decision efficiency and patient triage accuracy.

Led GenAI adoption efforts by building templates for healthcare summarization, entity extraction, and clinical knowledge retrieval.

Developed ETL workflows using AWS Glue, Lambda, Redshift, and S3 to combine EHR, claims, eligibility, and pharmacy datasets.

Engineered time-series forecasting models predicting admissions, utilization spikes, chronic-disease escalation, and resource needs.

Implemented anomaly detection for claims data, diagnosis patterns, outlier billing codes, and unusual utilization behaviors.

Built NLP models for medical NER, ICD/CPT extraction, note enrichment, and clinical semantic search.

Designed mapping documents, PHI field dictionaries, and validation rules supporting HIPAA-compliant AI pipelines.

Built MLflow/SageMaker-based model versioning, lineage tracking, monitoring dashboards, and automated retraining pipelines.

Applied SHAP/LIME explainability frameworks to produce clinician-friendly model explanations and feature insights.

Implemented drift monitoring, stability checks, and quality scoring for all deployed healthcare models.

Built automated validation scripts ensuring schema alignment, field completeness, mapping accuracy, and medical code consistency.

Built interactive clinician dashboards for risk scoring, care insights, GenAI summaries, and patient-level recommendations.

Conducted LLM prompt evaluation, hallucination analysis, and safety testing before production rollout.

Partnered with compliance, audit, and governance teams to ensure safe LLM usage and audit-ready documentation.

Collaborated with engineering and clinical informatics teams to continuously refine models based on provider feedback.

Kroger Data Engineer / ML Engineer February 2019 – August 2021

Built Spark ETL pipelines ingesting POS, supply chain, and merchandising datasets.

Developed Azure based curated datasets supporting forecasting, store operations, and executive dashboards.

Enabled ML driven demand-planning and promotion optimization initiatives across merchandising teams.

Partnered with retail leadership to identify data bottlenecks and redesign pipelines for operational accuracy.

Designed PySpark ingestion frameworks incorporating schema validation, deduplication, and transformation rules.

Built SQL based data-profiling and accuracy checks ensuring reliable upstream data feeds.

Optimized Spark workloads using partitioning, caching, and cluster configuration tuning.

Built dashboards for store KPIs, product availability, inventory alerts, and operational performance.

Developed feature pipelines for early ML models supporting SKU price optimization and sales prediction.

Created reusable ETL templates for ingestion across merchandising and supply chain workflows.

Integrated Delta Lake for versioned, ACID compliant curated retail datasets.

Performed root cause analysis of data gaps, schema mismatches, and ingestion inconsistencies.

Contributed to sprint planning, Agile ceremonies, and cross-functional engineering initiatives.

Supported debugging, schema updates, pipeline enhancements, and onboarding of new engineering team members.

Implemented automated quality scoring identifying missing values, delayed feeds, and erroneous transactions.

Built reconciliation scripts comparing transactional vs analytical outputs to ensure integrity.

Quark Software Inc, India Data Analyst / Junior Data Engineer June 2015 – November 2018

Developed Python and SQL scripts to extract, transform, and validate document metadata across high-volume content repositories.

Built ETL workflows to ingest PDFs, XML files, and publish content into internal processing pipelines.

Implemented metadata enrichment and document tagging automation to standardize content attributes.

Created feature extraction logic including text cleaning, tokenization, keyword extraction, and structural parsing.

Performed data quality checks, deduplication, and anomaly detection to ensure consistency across content datasets.

Designed dashboards tracking ingestion volume, processing errors, and workflow performance metrics.

Collaborated with engineering teams to troubleshoot schema inconsistencies and pipeline failures.

Authored data dictionaries, schema notes, and metadata mapping documents for internal engineering teams.

Participated in Agile ceremonies, sprint planning, and cross-functional collaboration for content engineering initiatives.

Contact this candidate