MS Computer Science student at New York University specializing in agentic AI systems and LLM orchestration.
2+ years of experience building production-grade AI systems with LangChain, LangGraph, PyTorch, and Hugging Face, deployed on AWS, Docker, and GCP. Skilled in end-to-end ML systems including fine-tuning, distributed deployment, and vector database optimization.
- 🔭 I’m currently working on GraphRAG memory and skills
- 🌱 I’m currently learning advanced multi-agent orchestration, LLM evaluation systems, and production-grade RAG architectures
- 💬 Ask me about agentic AI, LLM orchestration, RAG systems, transformers, and production ML pipelines
-
New York University, New York, NY
Master of Science in Computer Science, CGPA: 3.45/4.0
Sep 2024 - May 2026 (Expected) -
Birla Institute of Technology and Science, Pilani - Dubai, UAE
Bachelor of Engineering in Computer Science, CGPA: 3.5/4.0
Sep 2019 - Jul 2023
- Programming Languages & Frameworks: Python, SQL, JavaScript, TypeScript, Django, FastAPI
- ML/AI Frameworks: PyTorch, TensorFlow, Scikit-learn, Hugging Face, LangChain, LangGraph
- MLOps & Cloud: AWS, Kubernetes, Docker, PostgreSQL, MongoDB, Redis, Kafka
- Databases & Data Systems: MongoDB, PostgreSQL, Spark, BigQuery
- Web & Tools: React, Node.js, FastAPI, CI/CD Pipelines, Git, vLLM
-
Autonomous DevOps Copilot -
LangGraph,Gemini 1.5 Flash,Django,AWS SQS,PostgreSQL (pgvector),Angular- Dec 2025- Cut manual triage time by 30% by building an autonomous DevOps copilot processing 100+ daily GitHub/Slack alerts with multi-agent LLM workflows (LangGraph + Gemini).
- Scaled real-time event processing to ~5K+ events/min with <200ms latency using an AWS SQS-backed async architecture, improving system reliability under bursty webhook loads.
- Increased developer velocity by auto-generating code fixes and opening PRs via agent-driven CI/CD workflows; implemented human-in-the-loop approvals, achieving high merge acceptance rates (~70-80%).
- Engineered persistent agent memory and semantic context retrieval using pgvector (Neon PostgreSQL), reducing decision latency by ~40% and enabling context-aware automation via a real-time Angular dashboard.
-
CityLens -
Gemini Live,FastAPI,Vite,React Native,Google Maps API,Firestore,GCP- Jan 2026- Increased recruiter engagement by 25% by optimizing a React Native interface to a 98/100 Lighthouse score, improving performance and SEO.
- Built a FastAPI-based location intelligence system integrating Google Maps APIs (places, geocoding, directions) with Firestore session context, enabling real-time navigation, visual assistance, and live environmental insights across multiple interaction modes.
-
Real-Time Financial Fraud Explainer -
LangChain,Kafka (AWS MSK),AWS Bedrock AgentCore,Lenses.io- New York, NY
2nd Place / 25 teams - Lenses.io Real-Time Data & AI Hackathon (Oct 2025)- Built a real-time fraud detection pipeline processing streaming transactions via Kafka (MSK), enabling instant detection of anomalous credit-card/PayPal activity.
- Designed a 3-agent (Detection-Context-Explainer) LLM system producing interpretable fraud insights, improving explainability of flagged events.
- Enabled low-latency stream observability and agent-triggered reasoning using Lenses.io MCP, supporting real-time anomaly propagation across pipelines.
-
AI-Powered Shopping Automation System -
Python,Browser Use,DeepL API- New York, NY, AI Tinkerer Hackathon (Nov 2025)- Automated end-to-end grocery purchasing by building an agent pipeline that ingests lists from Google Docs/Notion and executes checkout via browser automation (Instacart/Target).
- Processed multilingual inputs using DeepL API and structured extraction workflows, improving accuracy of item/quantity parsing across heterogeneous sources.
- Orchestrated modular agents (ingestion-translation-execution) with Manus, enabling secure, scalable automation with OAuth-based integrations.
- GitHub: github.com/VijayGottipati
- LinkedIn: vijay-gottipati
- Portfolio: vijaygottipati.vercel.app
- Email:
vg2571@nyu.edu