L

Lakshmi Praneetha K

About

Detail

Applied AI Engineer | AI/ML Engineer | Generative AI | LLMs | RAG | Agentic AI | LangChain | AWS Bedrock | Machine Learning
California, United States
Contact Lakshmi regarding: 
work
Full-time jobs

Timeline

work
Job
school
Education

Résumé

Jobs verified_user 0% verified
  • ServiceNow
    AI Engineer
    ServiceNow
    Jul 2024 - Current (2 years 4 months)
  • Target
    Data Scientist
    Target
    Jul 2024 - Current (2 years 4 months)
    • Engineered personalization models with PyTorch and SQL, improving recommendation accuracy and driving measurable revenue growth.
    • Deployed real-time APIs with FastAPI and Docker handling 10K+ requests under 200 ms latency, maintaining 99.95% uptime.
    • Automated AWS MLOps pipelines (SageMaker, Kubernetes, GitHub Actions), cutting deployment time from days to hours and lowering costs by 18%.
    • Fine-tuned LLMs for metadata automation, reducing manual tagging by 70% and boosting SEO-driven product visibility.
    • Implemented predictive monitoring with Prometheus and Grafana, reducing incident resolution by 40% and saving millions annually.
    • Collaborated with product and marketing teams in Agile sprints, embedding AI-d
  • Deloitte
    AI Engineer
    Deloitte
    Jan 2021 - Jan 2023 (2 years 1 month)
  • Tata Consultancy Servicess
    Data Scientist
    Tata Consultancy Servicess
    Jan 2021 - Jan 2023 (2 years 1 month)
    • Built fraud detection pipelines using PySpark, SQL, and XGBoost, cutting false positives and preventing multimillion-dollar financial losses.
    • Developed churn prediction models with classification techniques and behavioral analytics, improving retention across telecom and banking clients.
    • Migrated ML workflows to Azure ML and GCP Vertex AI, reducing infrastructure costs and accelerating experimentation by 3x.
    • Designed data reconciliation and anomaly detection frameworks with Airflow and Delta Lake, achieving 99.9% accuracy in healthcare claims processing.
    • Delivered dashboards in Tableau and Power BI that reduced reporting turnaround and enabled faster executive decision- making.
    • Deployed predictive risk a
  • Postman
    Machine Learning Engineer
    Postman
    Jan 2020 - Jun 2020 (6 months)
Education verified_user 0% verified
  • Concordia UniversityWisconsin
    Master of Science - MS, Computer Science Information Systems
    Concordia UniversityWisconsin
    Jan 2023 - Dec 2024 (2 years)
  • A
    Bachelor of Technology - BTech, Computer Science and Engineering
    Auroras Technological and Research Institute
    Jul 2018 - Aug 2022 (4 years 2 months)
Projects (professional or personal) verified_user 0% verified
  • L
    Log File Processing
    • Data Collection & Parsing – Gathered raw web server log files and wrote Python scripts to parse IPs, timestamps, URLs, and status codes into structured format.
    • Data Transformation – Cleaned and enriched logs (e.g., extracted hour/day info from timestamps) using Pandas for easier aggregation.
    • Storage & Indexing – Loaded structured logs into PostgreSQL for querying (and optionally Elasticsearch for fast searches/visualizations).
    • Analysis & Reporting – Generated reports on traffic patterns, busiest hours, top pages, and error rates; visualized results using Matplotlib/Kibana.
  • G
    Generative AI–Powered Document Q&A System
    • Developed a Retrieval-Augmented Generation system using Hugging Face transformers, FAISS vector search, and LangChain to enable contextual question answering over large collections of unstructured documents.
    • Fine-tuned a LLaMA-based language model using LoRA techniques to improve response relevance and factual grounding, achieving approximately 30% improvement based on manual evaluation and retrieval accuracy checks.
    • Integrated the RAG pipeline into a FastAPI-based web application, automating document search and question answering workflows and significantly reducing manual lookup time for end users.
  • C
    CSV to Database ETL Pipeline
    • Data Ingestion & Cleaning – Collected CSV datasets (sales, customers, products) and cleaned missing values, duplicates, and inconsistent formats using Python and Pandas.
    • Schema Design – Designed relational database tables with proper keys and constraints (e.g., customer_id, product_id) to normalize and structure the data.
    • ETL Implementation – Built an ETL pipeline in Python to extract data from CSVs, transform into the required format, and load into PostgreSQL/SQLite.
    • Analytics & Dashboard – Wrote SQL queries to generate sales insights and built a simple Streamlit dashboard to visualize top- selling products and customer trends.
  • E
    End-to-End Predictive Maintenance System
    • Built a complete machine learning pipeline using Python, pandas, and XGBoost to predict equipment failures, applying structured preprocessing, feature engineering, and model evaluation techniques on historical sensor datasets.
    • Automated data cleaning, feature transformation, and validation workflows using pandas and scikit-learn, reducing manual preprocessing effort by approximately 60% and improving consistency across repeated training cycles.
    • Deployed the trained model as a real-time inference service using FastAPI and Docker, enabling faster failure detection and contributing to an estimated 25% reduction in equipment downtime.
This is a community-created genome.