CL

Charlie Linville

Founder & Builder — AI & Quantitative Systems

Right now I'm building AI and quantitative systems — agentic security research, location-aware assistants, and white-label creator platforms. Three products shipped in the last year. Co-founded a fintech startup before that. When I'm not shipping, I'm usually reading, making pour-over, or watching F1.


Journey

2026 — Present

Founder at Guleki

App publishing company focused on AI-native mobile experiences. Our first product, Geolocate Me, gives AI assistants real-time location context through a private GPS tracker with MCP integration — live on iOS and Android.

2026 — Present

Founder at Vectale

Building the future of creator growth and monetization.

2025 — 2026

CIO & Co-Founder at Agori

Early-stage fintech startup. Ran technology, product, and operations end-to-end. Architected and shipped the production platform (real-time quantitative paper trading competitions) across an 18-system stack on AWS. Built and led a team.

Production platform shipped in 14 weeks
2024 — Present

Wisconsin Quantum Computing Club

Participating in seminars on quantum computing and quantum information. Competed in the 2024 IBM Fall Fest hackathon.

🏆 IBM Fall Fest — Community Favorite + First to Finish
Oct 2024 — Dec 2025

UW-Madison Vice Chancellor Study Group

Selected through ASM Shared Governance to represent students. Collaborated monthly with the Vice Chancellor for Finance and Administration and university stakeholders on finance and administration topics.

2023 — 2024

Technology Improvement Consultant — Backpack Society

Architected a system solution and proposal to modernize the website, communication, and storage for a nonprofit supporting 70+ schools with kids in unstable food situations. After board approval, managed budget and infrastructure improvements. Reported to the board of directors as a special advisor.

2020 — 2024

Cyberpatriots — Cisco Lead

Competed in a national cybersecurity competition run by the Air Force Association. Led Cisco networking for my school team and trained 2+ members each year. Hardened Ubuntu 22, Fedora 36, Windows 10, and Windows Server 2019 systems for competition.

🏅 12th Place Nationally / 5,000+ Teams

Education

University of Wisconsin-Madison

2024 — 2028

BS in Applied Math, Engineering, and Physics (AMEP). Coursework spanning Modern Physics, Linear Algebra, Calculus, Chemistry, and Programming.

University of New South Wales — Study Abroad

Jan — May 2026

Semester abroad at UNSW Kensington in Sydney, Australia. Coursework in Numerical Linear Algebra, Philosophy of AI, and Mathematical Computing.


Projects

Dragonfly

2026

Built an autonomous security testing engine with model-agnostic execution-time safeguards. Wraps 114 security testing tools behind a composable policy pipeline with scope enforcement, action gating, and kill switches. Provides adapters for MCP, CLI, and LangChain/LangGraph. Ran a scripted ablation study (45 runs) and an LLM behavioral study (60 runs) to evaluate safeguard effectiveness. Includes 1,259 tests and a Docker test range.

PythonLangChainLangGraphMCPDockerGemini APIpytest

Vectale

2026

Built a multi-tenant SaaS platform for content creators to manage and deliver branded content to subscribers. Creators get a white-label portal with courses, news, links, and custom pages — all behind tiered subscriptions powered by Stripe Connect. Architected the full stack from domain routing and data isolation to CI/CD pipelines deploying across staging and production on DigitalOcean.

ReactTypeScriptNode.jsExpressPostgreSQLStripe ConnectCaddyTerraformDockerGitHub Actions

AI Powered Auction Arbitrage Agent

2025

Designed and deployed a system that aggregates and evaluates purchasing opportunities in auction markets. Uses a Python analytics pipeline to ingest listings, then performs GPU-accelerated OCR/computer vision and ML matching with an embedding engine to get market comps from a range of marketplaces. Persists results in AWS (DynamoDB, S3) for back testing and future neural network training.

PythonSeleniumEasyOCROpenCVeBay APIS3DynamoDBNumPyPandasOpenAI API

Agent Academy

2025

Built an automated evolution system that improves Claude Code agents using genetic algorithms. Treats agent prompts as a population of candidates, tests them against programming benchmarks, scores them with a multi-persona jury of specialized evaluators, and breeds the next generation through crossover and mutation. Runs unattended overnight with checkpoint/resume support.

PythonClaude CodeSQLiteTyperRichGenetic Algorithms

Covered-Call Optimization Engine

2025

Built a covered-call optimization engine that streams live options/price data from Interactive Brokers (IBKR) and recommends strike/expiration and position sizing under capital constraints. Implemented a Monte Carlo risk model to evaluate thousands of scenarios and compute expected P&L, win rate, and Value-at-Risk.

PythonClaude CodeNumPyPandasIBKR

Rydberg Blockade Gate Simulation & Pulse Optimization

2025

Simulated neutral-atom qubits in QuTiP: built time-dependent Hamiltonians for ground-to-Rydberg drives with detuning, modeled open-system dynamics via Lindblad operators. Designed and optimized Gaussian control pulses with SciPy to realize single- and two-qubit gates. Produced reproducible Jupyter analyses with parameter sweeps and clear plots of Rabi oscillations, coherence, and state populations.

PythonJupyterQuTiPNumPySciPyMatplotlib

Also Built

Tennis Alpha Finder

Quantitative in-play tennis betting system exploiting systematic mispricing on Kalshi. Markov chain model trained on 10,972 matches. 85.5% win rate across 152 paper trades.

Auto-Influencer

Multi-account X content automation platform. Ingests 650+ RSS feeds, generates persona-tailored posts with Claude, and schedules natively through the X API.

Idea Generation Engine

AI-powered market opportunity discovery engine processing 650+ RSS feeds. Clusters by similarity with TF-IDF and DBSCAN, surfaces opportunities via local 20B-parameter LLM.

Quantum Error Correction

Neural network for detecting local and non-local errors in quantum computers. Trained on 1M+ simulation runs, achieving 95%+ detection in a 12-qubit system.


Skills & Certifications

Languages

PythonJavaMatlabHTML/CSSBash/Shell

Technologies

AWSAgentic AIModel Context ProtocolPyTorchOpenCVDockerReact Native/ExpoQiskitCisco NetworkingLinux/UnixDynamoDBSQLSupabaseClaude APIOpenAI APILangChainOllama

Skills

Systems ArchitecturePrompt EngineeringAI-Assisted DevelopmentTechnical WritingCI/CDPenetration TestingLeadershipAgile

Certifications & Workshops

AWS Braket BadgeFoundational MatlabQiskit Global Summer SchoolUCLA 2025 Quantum Device Workshop

Misc

  • ☕ Specialty coffee — always dialing in the next shot
  • ⚛ Quantum computing — from Qiskit summer schools to Rydberg simulations
  • 🤖 AI systems — building agents, RAG, memory systems, running local models, prompt engineering
  • 🏁 F1 racing — the intersection of engineering and competition
  • 💪 Longevity training — optimizing for the long game

Writing

Read more on my Substack