Built internal AI tools for personalized workflows, alerts, and deadline tracking across eight concurrent automation workstreams while reporting to the CTO.
APR 2026 / PRESENT · REMOTEOPERATOR RECORD
I build AI software across inference, backends, and user-facing tools.
My recent work includes GPU model deployment, real-time VR prototypes, internal AI tools, distributed systems, and applied research.
Background
From low-level performance to the interface where someone uses the result.
I started professional engineering work at 16 and am completing a B.S. in Computer Science at UT Dallas, with an expected graduation date of May 2027.
I am most useful on projects that need both systems work and product judgment: measuring latency, connecting model output to a workflow, and deciding what must remain reviewable by a person.
Experience
Selected roles
Built a real-time prototype inside Meta Horizon OS that combined user input, environment signals, and interaction telemetry in a multimodal inference path.
JAN 2026 / APR 2026 · AUSTIN, TXWorked on XeSS 2.x model deployment and super-resolution shaders for game pipelines using C++, PyTorch, OpenVINO, HLSL, and DPC++.
MAY 2025 / AUG 2025 · AUSTIN, TXWorked on AI infrastructure integration and reliability under enterprise data and access constraints. Implementation details remain confidential.
JAN 2026 / PRESENT · REMOTEResearched model training, evaluation, and fine-tuning for mathematics and physics tasks.
MAY 2026 / PRESENT · DALLAS, TXDesigned asynchronous decision pipelines and low-latency data layers, then diagnosed Linux production bottlenecks across systems serving concurrent users.
AUG 2024 / MAY 2026 · SAN ANTONIO, TXCo-founded and built a financial tracking platform from zero through acquisition, including the application, data path, and deployment pipeline.
SEP 2024 / APR 2025 · BRONXVILLE, NYEducation and research
UT Dallas
B.S. Computer Science with a Business Administration minor. GPA 3.92. Expected May 2027.
Research interests
Model evaluation, persistent memory for agents, multimodal systems, and earlier medical-imaging work.
Technology
Tools grouped by type of use
“Regular use” and “working use” replace blanket skill ratings.
