Llama Lounge Snowflake Agentic Hackathon: BarackOllama, CrewAI, and the Golden Llama

Mar 1, 2026

Teaming up at Snowflake's Silicon Valley AI Hub to build BarackOllama—a CrewAI multi-agent pipeline leveraging Bright Data and Composio to generate role-tailored resumes and bespoke GitHub portfolios, earning entry into the coveted Llama Lounge.

AWS & NVIDIA Hackathon 2025: Research-as-a-Code, UDR, and the Big Cloud Bill

Nov 4, 2025

Diving into a month-long hackathon to build Research-as-a-Code—pitting NVIDIA Universal Deep Research against Google Test-Time Diffusion, wrestling with Llama-3.1-Nemotron-Nano NIMs on AWS EKS, and managing the realities of a hefty GPU Kubernetes cloud bill.

WeaveHacks 2: Building ¡alacard!, the Self-Improving AI Agent Cookbook with Google ADK and W&B Weave

Oct 12, 2025

Teaming up with Daniel Green and Amine Lbath at Weights & Biases HQ for WeaveHacks 2 to build ¡alacard!—an autonomous agentic factory that researches, composes, executes, traces, and self-heals Jupyter notebooks using Google ADK, W&B Weave OTEL telemetry, and Daytona sandboxes.

Gemma 3n Impact Challenge: Stigmergy ODEA, Edge AI Drone Swarms, and the Expensive YOLO Mode Doom Loop

Aug 6, 2025

Deploying multi-modal Gemma 3n on tiny Unihertz Jelly Star phones for autonomous disaster drone swarms, wrestling with Gradle and KMP, and surviving an expensive overnight AI agent doom loop.

AI Expo SCSP Hackathon 2025: Outbreak Ops and the Public Health Sentinel

Jun 6, 2025

Traveling to Washington DC to compete in the AI Expo SCSP Hackathon 2025 with Booz Allen Hamilton teammates Sydney Segear, Carlos J. Arguello, and Brianna Lynn. Building the Public Health Sentinel with LangGraph and Anthropic Claude, powered by overnight classical waltzes, and mastering native tool binding in agentic workflows.

LlamaCon 2025, Space Llama, and the Cerebral Valley Hackathon

May 5, 2025

Attending Meta's LlamaCon 2025 in San Francisco, seeing Booz Allen Hamilton's Space Llama featured in Mark Zuckerberg's keynote, discussing edge models at the LLM engineer lunch, and building LlamaLinked at the Cerebral Valley Hackathon at SHACK15.

Gemini API Developer Competition: Building OpenMMPA on Full Android Watches

Oct 19, 2024

Inspired by the AI pin craze and the Full Android Watch (FAW) movement, I built an open-source, multi-modal personal assistant with Flutter, Gemini 1.5, ObjectBox local vector RAG, and Cloud Functions for the Gemini API Developer Competition.

Chasing the Next Transformer Killer - Part 2

Jul 21, 2024

RWKV is one of the most underrated LLMs despite combining the advantages of RNNs and transformers

Chasing the Next Transformer Killer - Part 1

Jul 20, 2024

Short history of the Transformer architecture and some models that try to fix quadratic costs like Reformer, Performer, FlashAttention

No-Code Low-Code LLM Powered Quiz Game with FlutterFlow

May 30, 2024

Exploring what can be achieved with full-stack no-code solutions with my GDSC Fresno friends and another hackathon win