Brian Anderson

The latest advances in artificial intelligence (particularly large language models) continue to reverberate. Even for an “old school” AI person like myself (who cut his teeth with Prolog) it is clear that there has been a step change in our ability to create computer systems that can interact with humans using natural language. GPT-4 et al are exhibiting early signs of “common sense” and have encoded useful conceptual representations of the world. The debate rages on as to whether this is “intelligence”, but to an engineer like me, it sure seems useful! #LLM #AI #Intelligence #Programming https://lnkd.in/ebAgNapK

14

4 Comments

Given the widespread fame of generative AI, it's natural that interest in numerical optimization and solvers has waned. However, if a company is unable to use generative AI, utilizing various solvers can still be beneficial for enhancing the efficiency of project and program management activities. When I was at 3D Systems, I implemented a Linear Regression algorithm to estimate the printing time of 3D printers. The result was an accuracy of 99.75% for a 200-hour printing job. After transitioning from SQA to TPM, I primarily focused on project management methods. Recently, I realized that Linear Programming or Integer Programming can also be powerful tools for daily resource planning and engineer assignment. I found a great example from Google OR-Tools, which can be an excellent material if you're interested in using solvers for project management. However, this example doesn't account for each engineer's expertise. If given the opportunity, I will share how to set the objective/cost function to consider domain expertise. https://lnkd.in/g2XCwi-F

10

We released a new update of our Visual Studio Code Java extension. Go try it! Bunch of new improvements and already supporting JDK 23 Early Access builds as the extension is directly leveraging javac. https://lnkd.in/g8BKsNjS #java #IDE #OpenJDK #VSCode

61

1 Comment

Two exciting updates from Asana AI last week that you won't want to miss: 🌟 First, read about our latest work with Anthropic to test Claude 3.5 Sonnet and our process for testing new LLMs on our #engineering blog: https://lnkd.in/gJJfMta2 🌟 Second, my team is expanding, and I'm currently hiring! If you're interested, apply here: https://lnkd.in/grNWvw65 #AI #engineering #hiring #AsanaAI #teamexpansion #llms

14

According to Eric Schmidt, in the next year or so, the union of the large context window, agents, and text-to-action, when delivered at scale, is going to have an impact on the world that no one understands yet. Listen to the fascinating discussion in the "The Age of AI, Eric Schmidt" youtube video - https://lnkd.in/dFTKgBVY

52

1 Comment

In my latest Medium blog, I dive deep into how to use the Gemini-1.5-Flash-Pro-001 model to generate chapters for YouTube videos. We'll explore two methods: Method 1: Using the YouTube Transcription API and Gemini's text generation prowess to create chapters from very long transcripts given their 1M token context window. Method 2: Leveraging Gemini's multimodal capabilities to directly analyze YouTube videos (perfect for video owners!). Plus, I'll show you how to present these chapters in a user-friendly format (JSON, Dataframes) and even visualize them in a cool Gradio app. Ready to transform your YouTube experience? Check out the full blog post here: https://lnkd.in/gPCnMhMy #GenAI #YouTube #Gemini #GoogleCloud #AI

10