_       _                   __                  _    
   / \   __| | __ _ _ __ ___   | //  _   _  ___ ___| | __
  / _ \ / _` |/ _` | '_ ` _ \  |//| | | | |/ __/ _ \ |/ /
 / ___ \ (_| | (_| | | | | | | // |_| |_| | (_|  __/   < 
/_/   \_\__,_|\__,_|_| |_| |_| |_____\__,_|\___\___|_|\_\

Check This Out

The Latest Thing I've Put Together Hand Picked Just For You :)

How I Made A Deep Learning Robot

While AI is often associated with digital applications, its role in hardware and physical environments is just as exciting! To explore AI past software, I built and trained a robotic arm capable of learning and replicating tasks using a machine learning model called an Action Chunking Transformer (ACT). To do this, I 3D printed and wired both a "leader" and a "follower" arm, allowing me to teleoperate the follower and record demonstrations such as placing blocks into different buckets. Using this dataset, I trained the ACT model to learn from my examples, enabling the robotic arm to autonomously perform the task and even generalize to fixing its own mistakes.

What I'm Up To Now

Building AI Products and Teaching the World How to Use Them

Applied AI Engineer

2026 - Present

LangChain

Member of LangChain's Applied AI team, working on production agents and AI-related features.

AI Specialist

2023 - 2026

Cisco

Technical SME and hands-on developer for GenAI solutions on the MarTech Portfolio & Innovation Team, integrating AI marketing technology across the enterprise stack.

Director of Sales

2020 - 2022

The DTH Media Corp.

Led and trained a 15-rep advertising sales team, designing the commission structure, training program, and new ad products while managing local and national client relationships.

UNC Chapel Hill

2023

Degree in Economics, Statistics & Information Systems

Blog

Internal monologue

A Field Guide to Agent Evaluations

Jun 18, 2026

Practical taxonomy of agent evals: unit tests, integration tests, online evals, and benchmarks.

Claws… And the Rise of the 'Super Agent'

Apr 09, 2026

How the claw agent framework combines messaging, filesystems, and memory to create evolving digital assistants.

LLMs Out of Context

Feb 03, 2026

Context engineering, context rot, and how LLMs navigate content far exceeding their context window.

Agent Skills - Yet Another Tool Standard?

Dec 24, 2025

The new standard for packaging reusable workflow capabilities for filesystem-based agent harnesses.

How AI Engineers Improve Agentic Products

Dec 05, 2025

Defining, building, and applying LLM evaluations to improve AI products.

Reinforcement Learning with Verifiable Rewards for LLMs

Nov 03, 2025

Understanding RLVR and creating RL environments for large language models.

AI In Digital Worlds - Computer Use

Aug 15, 2025

LLMs should be given the same tools as humans to interact with the digital world we share.

Projects

My recent creations

See More
banana bench

Banana-Bench

Banana Bench is a benchmark for evaluating LLMs by pitting them head-to-head in a game of Bananagrams. LLMs must build valid crossword-style boards, demonstrating spatial reasoning, constraint satisfaction, and multi-turn strategic decision making in a competitive environment.

evaluizer

Evaluizer

Evaluizer is an interface for evaluating and optimizing LLM prompts. It allows you to visualize outputs against datasets, manually annotate results, and run automated evaluations using both LLM judges and deterministic functions. It features GEPA (Genetic-Pareto), an optimization engine that iteratively evolves your prompts to maximize evaluation scores through reflective feedback loops.

dca

DCA

Deep Competitive Analyst is a 'deep agent' style LLM assistant built to automate the creation of company profiles and competitive analyses. Built on top of LangGraph Platform and Perplexity Search, DCA can perform thorough research autonomously to create detailed business reports in a fraction of the time of a human. It operates for extended periods, dynamically spawning sub-agents to parallelize research tasks and creates the kind of in-depth competitive analysis that usually costs thousands.

rewrAIt

rewrAIt

Are you prompting the model or is the model prompting you? rewrAIt offers a unique 'text-editor' style interface for turn-based conversations with large language models that lets you revise any part of an LLM conversation- system, user, or AI messages, at any point in time. Add, modify, and remove context or change providers to your liking. Control the narrative with AI before it controls you.

SEB OCR

seb-ocr

seb-ocr is a custom built OCR and entity extraction pipeline for processing unstructured historical documents- part of a larger academic research project in the field of political science. Relying on LLMs to transcribe large volumes of scanned documents from official archives, then performing named entity recognition and extraction for downstream analysis.

QuicKB

QuicKB

QuicKB is an end-to-end machine learning pipeline that turns unstructured text into optimized, semantic knowledge bases with complimentary finetuned embedding models ready for RAG/AI retrieval. It combines the latest chunking approaches, synthetic training data generation, and dimension-reduced embedding model finetuning to create personalized domain-specific retrieval systems that are both more accurate and more efficient than generic methods.

Needle In A Vid Stack Logo

NIAVS

NeedleInAVidStack is a lightweight streamlit app that rapidly identifies, timestamps, and extracts specific content across large video and audio libraries. Rather than the tedious process of manually scrubbing through video and audio files yourself, NeedleInAVidStack uses Google's Gemini AI models to automatically and efficiently parse out exactly what you're looking for in a fraction of the time.

Project 1 Icon

ppt2desc

ppt2desc converts PowerPoint presentations into comprehensive machine-readable text formats using vision language models. It captures the full semantic meaning of slides by interpreting how text, graphics, and charts relate to each other- a crucial part of presentations that traditional scraping tools miss. Compatible with major AI platforms including OpenAI, Gemini, Anthropic, GCP Vertex AI, AWS Bedrock, and Azure AI Foundry.

More Content

Some of my other videos

Stop Prompt Engineering! Program Your LLMs with DSPy

Taking a deep dive into Stanford's declarative self-improving python (DSPy), showcasing how to program LLMs rather than rely on brittle text based prompting.

Knowledge Graph or Vector Database… Which is Better?

Combine RAG with knowledge graphs for richer insights. Use LLMs to extract entities and relationships, enabling structured reasoning and deeper context.

How I Made A Deep Learning Robot

Can GenAI interact with the real world? Absolutely! I trained a robotic arm using an action chunking transformer, enabling it to autonomously replicate and generalize tasks from teleoperated demonstrations.

AI Just Mastered 3D Design… What's next?

Discover how AI transforms text prompts into 3D models! From diffusion models to NeRFs, explore cutting-edge tech merging deep learning with 3D graphics.

BERT: The Most Used AI Model You Haven't Heard Of

While generative language models like GPT have captured public attention, BERT models remain the most widely deployed solution for production NLP tasks. Learn how BERT works and why it's so popular.

Create AI Images of YOU with FLUX Training

Open-source AI image generation now rivals enterprise solutions, offering flexibility and customizability. Learn how I trained FLUX.1 to create personalized images.

Wait... What REALLY Is A Vector Database?

Learn how vector databases store embeddings to capture meaning, enable semantic search, and power dynamic RAG systems for accurate LLM responses.

The BEST Way to Chunk Text for RAG

Optimize RAG systems with smarter text chunking. Explore strategies from basic splits to LLM-assisted chunking for better context and performance.

Make An AI Agent with OpenAI's Advanced Voice Mode

I tested OpenAI's Advanced Voice Mode to build low-latency voice assistants, integrating RAG to retrieve relevant context while maintaining natural, real-time spoken interactions.

Head over to my channel for more!