Tula Masterman

AI Solutions Leader & Applied Data Scientist

I explore what’s possible with AI and pioneer new approaches that solve important problems. I am endlessly curious, love a challenge, and am energized by working in an industry where today’s innovations redefine yesterday’s solutions. I’m passionate about technology and the endless possibilities it unlocks. I share what I learn to give back to the community that inspires me daily.

Skills & Experience

Generative AI & LLMs

AI Agents & RAG

Data Science & Analytics

Deep Learning (NLP & CV)

Python

AI Solution Architecture

Executive Advisory

Research & Technical Writing

Innovation Leadership

Public Speaking

For over five years, I've pursued my passion for data science and AI, including three years leading research and innovation initiatives. I've trained custom models for computer vision and natural language based tasks, and built generative AI solutions that pioneered innovative data pre-processing techniques and agent patterns using models from the world's leading providers like OpenAI, Meta, and Anthropic.

I've collaborated with incredible researchers, technical teams, and partners like Microsoft to deliver high-impact solutions that solve complex business challenges. I've published research papers, written articles frequently selected as Editor's Picks in Towards Data Science, and created educational videos that have reached 10k+ industry professionals.

I'm all in on everything I do -- whether it's swimming across the Catalina Channel in the middle of the night, reading 50+ books a year, or building compelling AI solutions. I value discipline, accountability, and fearlessness. I am always eager to learn and believe that openly sharing knowledge, debating, and challenging assumptions drive innovation and growth.

Videos

OpenAI Operator & What's Next For Computer Use Agents

How to have AI Agents with Super Powers

Understanding the DeepSeek R1 Hype

Airline Rebooking with AI Agents

Comparing AI models using Databrick's Playground

Exploring the Rise of Small Language Models

What can Dall-e 3 do?

Neudle, Skyrocket Your Business Data Search with AI

Research

Lightweight Safety Classification Using Pruned Language Models

This paper introduces Layer Enhanced Classification (LEC), a novel technique that outperforms GPT-4o and specialized models in content safety and prompt injection detection using fewer than 100 training examples with dramatically reduced computational requirements. The approach combines the computational efficiency of a streamlined Penalized Logistic Regression Classifier with the robust language understanding of an LLM. The results demonstrate that incredibly small transformer models (0.5B parameters) are robust feature extractors for classification tasks.

View on arXiv

The Landscape of Emerging AI Agent Architectures for Reasoning, Planning, and Tool Calling: A Survey

This comprehensive survey examines cutting-edge AI agent implementations, focusing on their capabilities for reasoning, planning, and tool execution. The paper maps out both single-agent and multi-agent architectures, identifies key design patterns, and provides critical insights for effective agent system development.

View on arXiv

Blog

Overcome Failing Document Ingestion & RAG Strategies with Agentic Knowledge Distillation

Improve document ingestion and information retrieval using the new Pyramid Approach

Improving Agent Systems & AI Reasoning

DeepSeek-R1, OpenAI o1 & o3, Test-Time Compute Scaling, Model Post-Training and the Transition to Reasoning Language Models (RLMs)

Introducing Layer Enhanced Classification (LEC)

A novel approach for lightweight safety classification using pruned language models

Computer Use and AI Agents: A New Paradigm for Screen Interaction

Exploring the future of multimodal AI Agents and the Impact of Screen Interaction

AI Agents: The Intersection of Tool Calling and Reasoning in Generative AI

Unpacking problem solving and tool-driven decision making in AI

Navigating the Latest GenAI Model Announcements - July 2024

A comprehensive guide to new models GPT4o-mini, Llama 3.1, Mistral NeMo 12B and other GenAI trends

Understanding Techniques for Solving GenAI Challenges

Dive into model pre-training, fine-tuning, RAG, prompt engineering, and more!

Are Language Models Benchmark Savants or Real-World Problem Solvers?

Evaluating the evolution and application of language models on real world tasks