+1 (401) 426-3005

Bhanu Teja Goske

Senior AI/ML Engineer | Generative AI & Intelligent Automation

Around 9 years designing and delivering production-grade ML, LLM and automation solutions across healthcare, life sciences, event technology and enterprise workflows

+1 (401) 426-3005
bhanutejagoske741@gmail.com
LinkedIn
GitHub

"Practical AI is about turning real business problems into reliable, production-ready intelligence."

Core Technologies

Python SQL PyTorch TensorFlow Azure AWS FastAPI REST APIs LLMs RAG LangChain LangGraph Azure OpenAI AWS Bedrock MLflow Docker Kubernetes FAISS Pinecone Tableau Power BI
Bhanu Teja Goske

About me

I'm Bhanu Teja Goske, a Senior AI/ML Engineer with around 9 years of experience designing, building, and supporting production-grade machine learning, Generative AI and intelligent automation solutions across healthcare, life sciences, event technology and large enterprise environments.

I specialize in end-to-end AI/ML systems — from problem framing, data preparation and feature engineering through model development, evaluation and deployment — and in building LLM-powered applications using prompt engineering, structured outputs, tool/function calling and Retrieval-Augmented Generation (RAG) with vector search, document chunking and domain-specific retrieval.

My work spans AI-driven and agentic workflow automation, scalable backend AI services and microservices in Python with FastAPI, and cloud-native deployments on Azure and AWS using containers, Kubernetes and CI/CD. I focus on solutions that balance model quality, engineering reliability and usability so that internal teams can adopt AI safely and effectively in production.

9+

Years Experience

71%

Alert Noise Reduction

3.4x

Fraud Detection Lift

65%

Incident Triage Automation

Want to learn more about me?

Chat with Astra, my AI-powered assistant, to explore my profile, experience and projects. All questions here are about me — not generic interview prep.

Astra AI

Bhanu's professional AI assistant · Online

Live
A

Hi! I'm Astra, Bhanu's assistant. Ask me anything about his certifications, technical skills, projects, or professional experience with each company.

For example: “What certifications does Bhanu hold?” or “What are Bhanu's main technical skills and focus areas?”.

Try asking Astra about Bhanu

These prompts are focused only on Bhanu's certifications, projects and detailed professional experience.

Technical Skills

Depth in Generative AI, LLMs, applied machine learning, production engineering and cloud-native deployment

Core Competencies

Generative AI • Large Language Models (LLMs) • Retrieval-Augmented Generation (RAG) • Prompt Engineering • Agentic AI • Tool/Function Calling • Structured Outputs • Semantic Search • Python-based AI Services • FastAPI & REST APIs • End-to-End ML Systems • Model Evaluation & Monitoring • MLOps & LLMOps • Cloud (Azure, AWS) • Vector Databases (FAISS, Chroma, Pinecone, Weaviate) • NLP & Deep Learning • LoRA & PEFT Fine-Tuning • Production-Grade Deployment & CI/CD

Generative AI, Agentic AI & NLP

Large Language Models (LLMs) RAG Prompt Engineering LangChain LangGraph Tool/Function Calling Structured Outputs Semantic Search Hugging Face Transformers OpenAI / Azure OpenAI Amazon Bedrock Sentence Transformers LoRA & PEFT MCP (Model Context Protocol)

Machine Learning & Deep Learning

PyTorch TensorFlow Keras scikit-learn XGBoost CNNs / RNNs / LSTMs Transfer Learning Named Entity Recognition (NER) Explainable AI (SHAP, LIME) Grad-CAM Model Evaluation Feature Engineering

Programming & Application Development

Python SQL JavaScript FastAPI Flask Python-based Microservices REST API Design Backend AI Services

MLOps, LLMOps & Deployment

Azure Machine Learning Azure OpenAI AWS SageMaker MLflow Kubeflow GitHub Actions Azure DevOps CI/CD Pipelines Docker Kubernetes (AKS, EKS) Weights & Biases LangSmith

Vector Search & Retrieval

FAISS ChromaDB Pinecone Weaviate Amazon OpenSearch Embeddings Document Chunking Retrieval Workflows

Mathematics & Statistics

Probability Statistics Linear Algebra Calculus Regression (Linear/Logistic) Hypothesis Testing Dimensionality Reduction A/B Testing Time Series Forecasting Evaluation Metrics (RMSE, MAE, F1, ROC-AUC)

Cloud Platforms

Microsoft Azure Azure OpenAI Azure Machine Learning Azure Databricks AKS Azure Data Factory Azure IoT Hub AWS SageMaker AWS Bedrock AWS Glue AWS Lambda Amazon S3 Redshift

Monitoring & Observability

Azure Monitor CloudWatch Grafana Logging & Tracing Production Monitoring

Analytics & Visualization

Tableau Power BI Streamlit Dash

Certifications

Professional certifications aligned with ML engineering and Generative AI delivery.

AWS Certified Machine Learning - Specialty

Anthropic AI Fluency

Microsoft Certified: Azure AI Engineer Associate

Certified Databricks GenAI Fundamental

Featured Projects

Production-oriented AI, Generative AI and intelligent automation solutions

Astra AI | Intelligent Voice & Workflow Automation Assistant

An AI-powered assistant designed for voice-driven workflow execution, intelligent automation and backend task orchestration across system and application-level actions.

  • Developed Python-based backend logic and API-integrated workflows for command execution, response generation and real-time interaction patterns
  • Implemented structured command handling and tool-triggered execution flows to support scalable AI-driven automation
  • Designed modular service integration patterns suitable for enterprise-style automation use cases
  • Focused on practical workflow execution, reliability and maintainable backend orchestration
Python FastAPI Voice Interfaces Tool/Function Calling Workflow Automation API Integration

Astra Guard | Real-Time Safety Monitoring and Computer Vision System

Developed a real-time AI safety monitoring system using computer vision to detect security threats, unsafe behavior, and suspicious activity in live video environments.

  • Built detection, alerting, and real-time inference pipelines to support surveillance and operational safety monitoring use cases
  • Implemented computer vision workflows using PyTorch, YOLO, and OpenCV for low-latency threat detection
YOLO Python OpenCV PyTorch Computer Vision Real-Time Inference Monitoring

CasePilot | AI-Assisted Legal Research and Decision Support Platform

Developed an LLM-powered platform for legal research, semantic retrieval, and decision support using retrieval-augmented generation, structured outputs, and domain-focused prompting.

  • Built legal search and summarization workflows to surface relevant case context
  • Implemented domain-specific RAG pipelines for grounded legal research and decision support
Python RAG Vector Search LangChain OpenAI API Structured Outputs Semantic Search

Professional Experience

Around 9 years building production-grade ML, Generative AI and intelligent automation solutions across healthcare, life sciences, event technology and enterprise environments.

Senior AI/ML Engineer

AT&T | Dallas, TX

May 2025 – Present

Leading development of a production-grade AI/ML and Generative AI platform supporting AT&T network operations with telemetry-driven fault detection, failure prediction, LLM-powered RCA, and controlled agentic automation.

Project: Intelligent Network Operations & Proactive Fault Management Platform

Led end-to-end platform development, reducing alert noise by 71%, lowering mean fault detection time from 45 minutes to 12 minutes, and enabling 65% automated incident triage.
Developed a PyTorch LSTM model for real-time temporal anomaly detection on streaming network telemetry data.
Built an XGBoost failure prediction model trained on 18 months of telemetry and incident data to forecast device failures up to 4 hours in advance.
Designed and implemented ML pipelines on Azure Databricks, including IoT Hub ingestion, SNMP/NetFlow feature engineering, model training, and deployment through Azure Machine Learning registry.
Built GPT-4 powered incident analysis workflows using prompt engineering, structured outputs and tool/function calling for grounded RCA and runbook-assisted troubleshooting.
Implemented RAG with LangChain and Pinecone over ServiceNow tickets, RCA reports, runbooks and operational documentation, plus hybrid retrieval and reranking.
Designed LangGraph agentic orchestration for alert triage, incident correlation, RCA generation and ServiceNow ticket drafting with human approval gates.
Implemented MCP-based integrations for enterprise tools including ServiceNow and Confluence, and deployed containerized services on AKS with Azure DevOps CI/CD.
Implemented observability and production monitoring using LangSmith, Azure Monitor and MLflow; mentored junior engineers on ML/GenAI engineering practices.
Python SQL PyTorch scikit-learn FastAPI REST APIs LLMs LangChain RAG Azure OpenAI Azure Machine Learning Azure DevOps AKS Azure Monitor AWS Docker MLflow

AI/ML Engineer

Cotiviti | Remote

Aug 2024 – Apr 2025

Contributed to an intelligent payment integrity and claims accuracy platform for U.S. health insurance payers, combining ML, NLP and Generative AI for audit workflow acceleration and higher payment accuracy.

Project: Intelligent Payment Integrity and Claims Accuracy Platform

Developed an XGBoost payment integrity scoring model with 180+ engineered claim/provider features, SMOTE for imbalance handling, and SHAP explainability.
Contributed to 3.4x higher confirmed fraud detection versus the legacy rule-based approach.
Built provider billing anomaly detection workflows using Isolation Forest and DBSCAN on longitudinal NPI-level billing patterns.
Fine-tuned BioClinicalBERT using PEFT/LoRA and implemented custom NER pipelines to extract medical entities and validate ICD-10/CPT coding accuracy.
Implemented RAG with LangChain, Pinecone and Azure OpenAI over CMS policies, coding guidelines and internal audit knowledge bases.
Built FastAPI-based services exposing payment integrity scores, NLP findings and AI case summaries; reduced per-claim review time from 22 minutes to under 6 minutes.
Supported HIPAA-compliant production readiness through Docker, Azure DevOps CI/CD, logging, monitoring and operational safeguards.
Helped drive an estimated $2.1M in incremental overpayment recovery per million claims.
Python SQL PyTorch scikit-learn FastAPI REST APIs NLP LLMs LangChain RAG Azure Machine Learning Azure DevOps AWS Docker MLflow

Machine Learning Engineer

Cvent | Tysons Corner, VA

Dec 2020 – Jul 2023

Built machine learning and AI-driven solutions that improved platform intelligence, personalization, attendee insights and content workflows across event technology products.

Project: Machine Learning Solutions for Event Platform Intelligence and Personalization

Built a hybrid session recommendation engine using collaborative and content-based filtering, increasing session click-through by 38%.
Developed a BERT-based sentiment analysis pipeline for attendee feedback across 5 dimensions, achieving 0.83 F1-score.
Designed an event ROI and attribution model using XGBoost and Salesforce engagement features for explainable business impact scoring.
Implemented retrieval-augmented content generation with LangChain, Pinecone and Azure OpenAI, reducing content creation time from 18 minutes to under 4 minutes.
Built and deployed production-grade ML/AI services with FastAPI, Docker, AKS and Azure DevOps CI/CD for low-latency inference during peak event activity.
Partnered with product, engineering and marketing teams to integrate ML outputs and support production readiness through debugging, documentation and optimization.
Python SQL scikit-learn PyTorch TensorFlow Pandas Feature Engineering Model Evaluation Docker Git

Data Scientist

Axtria | India

Mar 2018 – Nov 2020

Built life sciences commercial analytics and predictive models supporting HCP targeting, prescribing propensity, therapy adoption, territory design, marketing effectiveness and prescription forecasting.

Project: Life Sciences Commercial Analytics and Predictive Insights

Built HCP prescribing propensity models using XGBoost on longitudinal Rx/CRM activity data with SHAP explainability, achieving average AUC of 0.81.
Developed brand adoption and therapy switch prediction models using Logistic Regression and Random Forest, contributing to an 18% increase in sales call conversion.
Designed territory optimization models using linear programming to balance HCP opportunity, geographic alignment and rep workload.
Implemented Bayesian marketing mix models with adstock transformation to estimate incremental prescription lift by channel.
Built ARIMA and Prophet forecasting models for prescription volume prediction across short- and mid-range planning horizons.
Developed reusable Python/SQL pipelines for feature engineering, model validation and reporting automation across concurrent analytics engagements.
Python SQL scikit-learn Pandas Predictive Analytics Statistical Analysis Data Analysis Git

Python Developer

Medline | India

Mar 2016 – Feb 2018

Developed Python backend automation and data processing solutions for healthcare supply chain operations, product data workflows and internal reporting systems.

Project: Backend Automation and Data Processing for Healthcare Operations

Built Python ETL pipelines for large-scale medical product data processing with automated validation rules to improve data quality.
Developed REST integration connectors in Python to synchronize ERP, inventory and product catalog systems and reduce manual data handling.
Built automated reporting pipelines with Python/SQL for inventory scorecards, fulfillment summaries and product data quality reports.
Developed Flask-based backend APIs for internal product and inventory tools to improve retrieval performance and usability.
Implemented structured logging, exception handling and diagnostics to strengthen production support and troubleshooting.
Built AWS S3 archival workflows with boto3 and used Azure Blob Storage for cross-region sharing; designed reusable modular pipeline components.
Python SQL REST APIs Relational Databases Git Linux

Let's Connect

Ready to discuss AI/ML, Generative AI or intelligent automation opportunities and collaborations? Let's talk!

Get In Touch

Email

bhanutejagoske741@gmail.com

Phone

+1 (401) 426-3005

Location

United States

Availability

Available for new opportunities