logo
episode-header-image
Jul 3
40m 47s

Episode 246 - Reasoning About Gemini 2.5...

Mark and Allen
About this episode

Join Allen Firstenberg and Mark Tucker as they dive into Google's latest Gemini 2.5 models and their much-touted "thinking" capabilities. In this episode, they explore whether these models are genuinely reasoning or just executing sophisticated pattern matching. Through live tests in Google's AI Studio, they pit the Pro, Flash, and Flash-Lite models against tricky riddles, analyzing the "thought process" behind the answers. The discussion also covers the practical implications for developers, the challenges of implementing these features in frameworks like LangChainJS, and the broader question of what this means for the future of AI.


[00:00:00] - Introduction to Gemini 2.5 "thinking" models

[00:01:00] - How "thinking" models relate to Chain of Thought prompting

[00:03:00] - Advantages of separating reasoning from the answer

[00:05:00] - Exploring the models (Pro, Flash, Flash-Lite) in AI Studio

[00:06:00] - Thinking mode and thinking budget explained

[00:09:00] - Test 1: Strawberry vs. Triangle

[00:15:00] - Test 2: The "bricks vs. feathers" riddle with a twist

[00:17:00] - Prompting the model to ask clarifying questions

[00:25:00] - Is it reasoning or just pattern matching?

[00:28:00] - Practical applications and the future of these models

[00:35:00] - Implementing reasoning models in LangChainJS

[00:40:00] - Conclusion


#AI #GoogleGemini #ReasoningModels #ThinkingModels #LLM #ArtificialIntelligence #MachineLearning #LangChain #Developer #Podcast #TechTalk #TwoVoiceDevs

Up next
Jun 26
Episode 245 - From Python to TypeScript: Coding JCrew AI to Build Better Agents
Ever find that the best way to understand a new framework is to build it yourself? In this episode of Two Voice Devs, Mark Tucker takes us on a deep dive into Crew AI, a powerful Python framework for orchestrating multi-agent AI systems.To truly get under the hood, Mark decided t ... Show More
33m 18s
Jun 20
Episode 244 - What's New With Anthropic?
What do Anthropic's latest announcements mean for developers? In this episode, Allen is joined by freelance conversation designer Valentina Adami to break down all the major news from the recent "Code with Claude" event.Valentina shares her hands-on experience and perspective on ... Show More
34m 28s
Jun 12
Episode 243 - AI Agents: Exploits, Ethics, and the Perils of Over-Permissive Tools
Join Allen Firstenberg and Michal Stanislawek in this thought-provoking episode of Two Voice Devs as they unpack two recent LinkedIn posts by Michal that reveal critical insights into the security and ethical challenges of modern AI agents.The discussion kicks off with a deep div ... Show More
30m 57s
Recommended Episodes
Nov 2024
Making Sense of Agentic AI | ThoughtWorks Birgitta Boeckeler
There’s AI agents. There’s AI tooling. Do either drive business impact or are they just more things your dev team is supposed to stay on top of? Birgitta Boeckeler, Global Lead for AI Assisted Software Delivery at ThoughtWorks, joins the show to discuss the practical applications ... Show More
47m 40s
Sep 2023
Meta’s Quest 3, AI chatbots and Ray-Ban smart glasses
This week, it’s Meta’s turn to highlight AI during its device event. In this episode, Devindra and Cherlynn dive into all of the news from Meta’s Connect 2023 event, where it unveiled Meta AI and accompanying celebrity-powered chatbots. Oh yah, and it introduced the Meta Quest 3 ... Show More
1h 6m
Sep 2024
Study Reveals Vulnerabilities in Alexa, Siri, and Google Assistant to Malicious Commands
In this episode, we explore a recent study that uncovers how popular voice assistants like Alexa, Siri, and Google Assistant are susceptible to malicious commands. We discuss the potential risks and what users can do to protect their devices. Get on the AI Box Waitlist: ⁠⁠⁠https: ... Show More
6m 17s
Nov 2024
SN 1001: Artificial General Intelligence (AGI) - Gmail Temp Addresses, Russia's Internet Off Switch
How Microsoft lured the US Government into a far deeper and expensive dependency upon its cybersecurity solutions. Gmail to offer native throwaway email aliases like Apple and Mozilla. Russia to ban several additional hosting companies and give its big Internet disconnect switch ... Show More
2h 26m
Sep 2024
AI is more than GenAI
GenAI is often what people think of when someone mentions AI. However, AI is much more. In this episode, Daniel breaks down a history of developments in data science, machine learning, AI, and GenAI in this episode to give listeners a better mental model. Don’t miss this one if y ... Show More
40m 3s
Jan 2021
How Salesforce will make Einstein smarter in 2021
Salesforce launched Einstein, its artificial intelligence tool, in 2016. It was memorable because of the marketing materials, featuring a cute cartoon of the world's most misquoted-scientist. It was also memorable because of the unique capabilities Einsten brought to the table. T ... Show More
27m 46s
May 2021
397: Customer Feedback vs. Team Intuition
This week, we talk about the tension between building what customers explicitly ask for versus building towards a team’s internal vision. In The Sidebar, we talk about the lack of public software critique: Why isn’t there an MKBHD equivalent for software design?Golden Ratio Suppo ... Show More
23m 29s
Nov 2024
How AI is changing national security w/ Kathleen Fisher
We’ve had conversations about AI’s online influence on politics, from deepfakes to misinformation. But AI can also have profound effects on hardware – especially when it comes to national security and military capabilities like weapons and stealth technologies. Kathleen Fisher is ... Show More
55m 1s