José Luis Olave Dañobeitia’s Post

Explore how live programming intersects with literate, local-first, and functional programming—and its potential to shape the future of AI software. 🌐 F5's Gilad Bracha looks at how a humanistic approach to computing can preserve individual control in the AI era. 🤖 http://ms.spr.ly/6043lqij1

Explore how live programming intersects with literate, local-first, and functional programming—and its potential to shape the future of AI software. 🌐 F5's Gilad Bracha looks at how a humanistic approach to computing can preserve individual control in the AI era. 🤖 http://ms.spr.ly/6049lsbl1

Explore how live programming intersects with literate, local-first, and functional programming—and its potential to shape the future of AI software. 🌐 F5's Gilad Bracha looks at how a humanistic approach to computing can preserve individual control in the AI era. 🤖 http://ms.spr.ly/6041lIgd1

Future of Coding: Best AI Code Assistants to Watch Out for in 2025 Best AI Code Assistants of 2025 A new era has dawned in software development. Artificial Intelligence is constantly changing the ways code is being written, tes https://lnkd.in/eQi7dumU

Future of Coding: Best AI Code Assistants to Watch Out for in 2025 Best AI Code Assistants of 2025 A new era has dawned in software development. Artificial Intelligence is constantly changing the ways code is being written, tes https://lnkd.in/eQi7dumU

DeepSeek-V3 is now available on Poe! DeepSeek-V3 stands out as one of the most powerful open-source models available and excels in math, coding, and reasoning tasks. Its long context handling and performance across key benchmarks rival leading closed-source models. It is hosted by Fireworks AI and available for you to try at https://lnkd.in/g984Wec3 and all Poe apps.

Simple coding tasks can be done using Anthropic's "Computer use" and Claude. It's a bit rough around the edges, but it gets you there. Just sit back a few months and see how tools like this become mature in your field too. #exponentialGrowth

👨💻 Why Claude 3.5 Sonnet Outperforms o1 Preview in Coding Tasks⁉ OpenAI's o1 Preview vs Anthropic's Claude 3.5 Sonnet differ notably in coding performance. Despite o1 Preview's advanced reasoning (CoT, "Chain-of-Thought"), Claude 3.5 Sonnet excels in coding benchmarks!, scoring 90.4% on the MMLU test. It also supports a larger input context of 200K tokens, compared to o1 Preview's 128K. Interested...? For a detailed comparison, official analysis: https://lnkd.in/d7JPNBaf https://lnkd.in/d5PKMqkc

From Code to Autonomous Agents: A 6-Step Spectrum for Building Applications in GenAI era: 1️⃣ Code – Traditional programming for precise logic. 2️⃣ LLM Call – A single prompt and response to one model. 3️⃣ Chain – String together multiple LLM outputs in sequence. 4️⃣ Router – Direct tasks to different models without loops. 5️⃣ State Machine – Allows cycles and complex branching. 6️⃣ Agent – Autonomous decision-maker orchestrating multiple steps. As you move from Code to Agent, you trade control for flexibility. Which level fits your project best?

I'd like to share what's on my mind about technology, software, and us humans. First of many to come--I hope this article finds you curious. #SoftwareDevelopment #TechHumanity #AI #TechTeams #Collaboration