Jyoti Das

About Jyoti Das

Resume: https://drive.google.com/file/d/185Gj1qajP9-_OEsRid47bASPu-zIAsfp/view?usp=sharing 13 years of hands-on Experience, delivering end-to-end Gen AI/Machine Learning & Data Science solutions; strong storytelling skills with a track record across startups & Fortune 150 companies - Tech, Finance, Energy/Utilities, Insurance, Telecom, Logistics. Delivered projects; managed clients in cross functional business units – Marketing, Sales, Product, Operations, Engineering Mentor for Great Learning EdTech platform (partnership with UT Austin); teach PG AI/ML program. Speaker at Ai4 & REWORK AI conference: Gaussian Mixture model application in a business problem Led technical projects; Managed & mentored teams of 1–5 across multiple initiatives; acted as strategic advisor Hands on AI/Machine & Deep Learning, Generative AI models development (Python, Tensorflow, PyTorch): Time Series Forecasting, NLP, Chatbot, Propensity, Clustering, Recommendation; worked in Azure Databricks, AWS (Sagemaker, Jumpstart, Bedrock). Collaborated; managed consultants/architects from leading firms: Microsoft, Databricks, SAP, E&Y, Accenture, Syntelli solution. Developed risk-based Propensity model for DNP natural gas customers; built ML/NLP algorithms for intelligent Grid analytics platform, Energy Efficiency dept at Duke Energy. Deployed Propensity models to SAP Big Data Services cloud with architects, automated & built ETL pipeline, data engineering (PySpark/Spark SQL). Production model generated -$240MM in incremental annual revenue.https://github.com/dipjyotidasSkills:Software : Python (keras, scikit-learn, networkx), AWS (Bedrock, Sagemaker, OpenSearch), Azure Databricks, GCP Vertex AI, Kubernetes, Docker, Microservices, Snowflake, PostgreSQL, Argo workflow, Azure DevOps, Github actions, CircleCI, Elastic Search, FastAPI, Flask, Pyspark, Streamlit, Tensorflow, C++Machine/Deep Learning/Model Serving: Regression (Lasso, Ridge), Classification (Logistic regression, Random Forest, KNN, Naive bayes), Boosting (XGboost), Clustering (Gaussian mixture, DBSCAN, K-means), Graph Network, RNN, LSTM, CNN, Encoder-Decoder, MLOps, CI/CD/CT, Databricks Model registry, MLflow, TeamCityGenerative AI: GANs, Transformer, BERT; LLM-GPT-4o, Gemini, Llama 3, Mistral, Claude, Mistral, LangChain, Llamaindex, LangGraph, Google Agent Dev Kit, CrewAI, FastMCP, A2A protocol, Langfuse, Embeddings – Hugging face, Cohere; Vector DB – Vespa, OpenSearch; Guardrails, RAGAs, Open AI Moderation API, NLP, Word2vec, RL- RLlib, Gymnasium

Experience

1. Gen Ai Architect

   Ascendion

   Oct 2025 — Present · Charlotte, NC, US

   Ascendion is a leader in AI-powered software engineering - it has developed AAVA Agentic AI platform, enabling automation across software development lifecycle. The company implements Gen AI projects for multiple enterprise clients. Lead and manage a cross-functional team of AI engineers, software developers, data scientist, UI/UX designer for Charter communications. I’m leading the architecture & development of a unified Agentic AI chatbot service within Charter’s Tech Mobile application, used by field technicians, supervisors, site maintenance, construction engineers. Built with Langgraph, orchestrator agent identifies role and intent of user query; coordinates task-specific agents (Job Insight, Tech Assist, APIs); integrated Langfuse for observability. Agentic chatbot service delivers instant, role-aware assistance - reducing 60% support calls, accelerating service completion & annual savings of millions of dollars. Leading the development of RAG service by extracting, embedding over confluence pages for site construction engineer maintaining page hierarchy, context in various levels, attachments – images, pdfs, excel, word with PGVector DB. Responsible for end-to-end solution architecture, client demos, AWS EKS deployment, CI/CD pipeline for microservices, and infrastructure management; perform code reviews, guide technical design, and lead hands-on POCs. Collaborate closely with clients on delivery management, roadmap planning, KPI tracking, business use case definition; advise stakeholders on build/buy/rent strategies, LLM and embedding model selection, and vector database/observability tools.

Education

• University of Florida

  Masters of Science, Materials Science and Engineering

  2011 — 2012

• National Institute of Technology, Tiruchirappalli

  Bachelor of Technology, Metallurgical and Materials Engineering

  2006 — 2010

Skills

• Windows
• Hadoop
• Report Writing
• Scanning Electron Microscope
• Microsoft Office
• Visual Basic
• Autocad
• Html
• Pareto Analysis
• Atomic Force Microscope
• R
• Non Destructive Evaluation
• C++
• Afm
• Characterization
• Apache Spark
• Tableau
• Anova
• Statistical Modeling
• Time Series Analysis
• Aws Ec2
• Statistical Process Control (Spc)
• Synthesis of Materials
• Labview 2011
• Ellipsometer
• Machine Learning
• Design Expert
• Adobe Acrobat
• Forecasting
• Outlook
• Minitab
• Electron Beam Evaporator
• Technical Presentations
• Quality Control
• Profilometer
• Excel
• Microsoft Sql Server
• Scala
• Visual Basic for Applications (Vba)
• Characterization of Materials