Skip to content
View ghantapavan93's full-sized avatar

Block or report ghantapavan93

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
ghantapavan93/README.md

Pavankalyan Ghanta

LinkedIn β€’ Email β€’ Portfolio β€’ Resume (PDF)


βœ” Hands-On Expertise in Agentic Artificial Intelligence (AI), Retrieval-Augmented Generation (RAG) pipelines, and cloud-native microservices
βœ” Proven track record of building production-ready AI systems across healthcare, surveillance, education, and enterprise automation
βœ” Skilled at turning complex AI workflows into scalable, human-friendly systems (Large Language Model Operations (LLMOps), Langfuse, OpenTelemetry)
βœ” Experienced in real-time inference, cross-domain agent collaboration, and cloud-native deployment (Amazon Web Services (AWS), Google Cloud Platform (GCP), Vertex AI)
βœ” Startup agility + enterprise discipline β€” from fast minimum viable product (MVP) cycles to enterprise-grade deployments at University of North Texas (UNT) Research Labs


I’m an AI Engineer and Full-Stack Product Builder who turns real-world pain points into deployed systems people actually rely on β€” not prototypes that only look good in slides. I care about one thing: reducing friction in moments that matter β€” helping someone understand their health, keeping crowds safer in real time, or giving teams faster answers they can trust.

I build the full stack end-to-end: application programming interfaces (APIs), internal tools, AI copilots, and cloud-native platforms with reliability and observability as first-class features. Across work with Krowd Guide, University of North Texas (UNT) Research, and Builder.ai, I’ve shipped agentic workflows (Planner β†’ Retriever β†’ Critic), hardened RAG pipelines, and production microservices instrumented with Langfuse + OpenTelemetry β€” so performance, failures, and fallbacks are visible and fixable under real load.

Right now, I’m building LangGraph-based agents with policy guardrails, tool-use, and hybrid RAG + graph retrieval for cross-domain intelligence β€” systems that stay grounded, auditable, and operational when the data is messy and the stakes are real.


πŸš€ Production systems I’ve shipped:
🩺 Medical AI copilot helping patients decode complex diagnoses into plain language
πŸ‘₯ Real-time crowd safety monitoring for large events with live detection + operator dashboards
πŸ§‘β€πŸ’Ό Human Resources (HR) automation tools saving teams 10+ hours/week on admin workflows
πŸ” RAG-powered knowledge engines for enterprise teams with high-confidence answers + explainability

β€œI build systems that fit how companies already operateβ€”APIs, internal tools, AI copilots, and platforms that handle real edge cases and are actually used by real people. Not just demos that look good in slides.”


❀️ HealthTech & Wellness β€’ πŸ₯ Healthcare & Clinical AI β€’ πŸ” Cybersecurity & Threat Intelligence β€’ πŸ‘₯ HR Tech & People Operations (PeopleOps)
πŸ“š Education Technology (EdTech) & Learning β€’ β™Ώ Accessibility & Inclusive Design β€’ ✈️ Travel / Transit / Events β€’ πŸ’³ Finance / Payments / Pricing
πŸ—ΊοΈ Real Estate & Location Intelligence β€’ πŸ› οΈ Developer Platforms & Internal Tools β€’ πŸ—οΈ Engineering & Infrastructure β€’ 🌍 Open to Applied AI in new domains


πŸ€– AI Copilots & Assistants β€’ πŸ“š RAG Systems & Knowledge Engines β€’ πŸ“Š Real-Time Dashboards & Analytics
🧩 Internal Tools & Admin Portals β€’ 🧠 Machine Learning (ML) / Large Language Model (LLM)-powered APIs & services β€’ πŸš€ Full-Stack MVPs & Platforms


AI-native cyber intelligence platform that converts internet-facing host findings into structured, operator-ready risk summaries using deterministic validation + RAG and production guardrails.
Stack: FastAPI β€’ React + TypeScript β€’ Vector Search β€’ Schema Validation β€’ OpenTelemetry β€’ Langfuse

Clinical workflow-oriented platform focused on reducing healthcare friction: patient records organization, coordination flows, and fast access to critical context through a clean full-stack experience.
Stack: Full-Stack Web App β€’ Secure API-first design β€’ Workflow-driven user experience (UX)

Domain-specific RAG system resolving clinical queries with Facebook AI Similarity Search (FAISS) + Pinecone; Graphics Processing Unit (GPU)-accelerated on AWS Elastic Compute Cloud (EC2) using Compute Unified Device Architecture (CUDA).

Real-time surveillance intelligence detecting human actions & summarizing Closed-Circuit Television (CCTV) footage via Apache Kafka + GPT-4.

Multi-agent tutor with LangChain + ChromaDB + MiniLM, built for onboarding & training support.

Personalized wellness coach integrating wearable APIs + RAG orchestration; deployed via AWS Lambda.


Languages

AI & Generative Artificial Intelligence (GenAI)

Cloud & Infrastructure

Backend & Full-Stack Development

Cloud & Infrastructure

Data Engineering & Analytics

Version Control & Collaboration


πŸ“₯ Download My Resume (PDF)



Pinned Loading

  1. Censys-Summarization-Agent Censys-Summarization-Agent Public

    Python 1

  2. Smart-Tutor-AI-Clean Smart-Tutor-AI-Clean Public

    Jupyter Notebook

  3. project-h project-h Public

    Forked from aftabhshaik/project-h

    www.projhealth.com

    JavaScript

  4. Eagle-Eye-AI Eagle-Eye-AI Public

    Python

  5. 100MoS-Proto 100MoS-Proto Public

    TypeScript

  6. EPISODE-COMPANION-AGENT EPISODE-COMPANION-AGENT Public

    Episode Companion Agent: multi-persona RAG microservice powering interactive AI podcasts

    Python