Praneeth Ravuri
Building autonomous agents and distributed systems for the real world.
I'm an AI Engineer focused on production-grade agentic AI, backend infrastructure, and high-throughput data pipelines. Currently at Tuskira, building Agentic AI and scalable systems for cybersecurity.
Experience
Tuskira
AI EngineerDec 2025 — Present · United States
Building autonomous AI agents for cybersecurity investigations that query many data sources and hold context across long sessions. Designed a memory layer that separates long-term knowledge from active reasoning, storing summarized investigations in Neo4j and Pinecone. Optimized MCP servers in Go and Python to filter data before the LLM, cutting token usage. Deployed on AWS with Kubernetes across multiple client environments.
Lumen
Software EngineerOct 2024 — Nov 2025 · United States
Built backend services processing millions of network flows per minute for a traffic-analysis platform. Architected an event-driven Kafka pipeline that decouples ingestion from processing so failures can’t cascade. Added backpressure in Go with buffered channels to absorb spikes and hold throughput at peak. Shipped on Docker and Kubernetes with zero data loss in production.
ADP
Full Stack Engineer InternJan 2022 — Jun 2022 · India
Modernized a legacy employee portal serving thousands of daily users. Built a React frontend and added a Redis caching layer to the Node.js backend to cut repeated database queries. Improved response times and stabilized performance during peak traffic.
Projects
Gary
Agentic AI · LLMsAn AI agent that rewrites résumés to match job descriptions without keyword stuffing. A multi-step Crew AI workflow extracts role requirements, analyzes tone, and rewrites experience while keeping the narrative consistent — outputting tailored PDF/Word docs that pass ATS filters.
Pitstop
MCP · Streaming DataAn MCP server connecting live F1 telemetry to LLMs for natural-language race queries. Processes high-frequency streams with FastMCP and HttpX, filtering noise into signals like lap times, tire degradation, and pit windows — enabling conversational queries about strategy and pace.
Smart Traffic
Reinforcement LearningA reinforcement-learning simulation using SARSA to optimize traffic-light timing from real-time queue lengths. Modeled intersection dynamics with NumPy and Pygame, training an agent to learn adaptive switching policies instead of fixed schedules — reducing average wait times versus static timers.
Notstuck
RAG · Hybrid SearchA hybrid-search RAG system combining vector similarity and keyword filtering for technical-doc retrieval. Built with a Next.js + React frontend, PostgreSQL for metadata, and Pinecone for vectors. Crew AI generates cited answers, balancing semantic understanding with exact-term matching.
Education
Master’s in Computer Science
George Mason University
Bachelor’s in Computer Science
GRIET