🧠 How to use AI to prep, code, and create faster

🔥🔥🔥 Three big updates

1. OpenAI launches ChatGPT Images (GPT Image 1.5)

OpenAI rolled out a major upgrade to its visual AI features with GPT Image 1.5, now integrated as a new Images tab in ChatGPT and available via API.

The model generates visuals up to four times faster than before, follows complex instructions more closely, and lets you make targeted edits — like adding, removing, or combining elements — while preserving fine details and composition. It also handles small, dense text with greater fidelity and introduces visual workspace features meant to feel more like a creative studio than a prompt box.

Early feedback highlights practical editing control and speed as big wins, though many still compare quality head-to-head with tools like Google’s Nano Banana Pro.

2. Zencoder launches Zenflow for AI-assisted coding

Zencoder introduced Zenflow, a free desktop platform designed to help engineers build better software using AI. Rather than relying on ad-hoc prompts, Zenflow lets you define structured, spec-driven workflows where multiple AI agents coordinate to plan, implement, test, and verify code in parallel.

This shifts the focus from “guessing prompts” to repeatable, reliable engineering practices with automated validation built in — a real step up for teams pushing AI toward production-ready development. 

Developers appreciate this shift because it introduces built-in verification and multi-model coordination, rather than ad-hoc outputs that can diverge from the intended logic.

3. Microsoft’s TRELLIS 2 advances image-to-3D generation

Microsoft Research unveiled TRELLIS 2, a state-of-the-art generative model for turning images into high-fidelity 3D assets. With 4 billion parameters and a novel sparse voxel representation called O-Voxel, TRELLIS 2 can handle complex shapes, topologies, and textures in a way previous 3D generators struggled with.

It reconstructs meshes with photorealistic details and materials from a single image input — a leap forward for creators, designers, and game developers who need realistic 3D content fast.

🔥🔥 Two tools worth trying

Google Labs CC
A new conversational AI agent from Google Labs that lets you automate tasks, pull context from your content, and manage workflows conversationally. It’s designed to feel like an AI collaborator rather than a traditional script engine, helping you break down tasks, iterate on drafts, or explore ideas without constant menu diving.

Okara
A unified interface for working with 30+ open-source AI models, so you can switch between capabilities (vision, text, search, analysis) without wrangling infrastructure. Okara also lets you query web sources like Google, Reddit, YouTube, and more directly from chat, making research and brainstorming much faster for teams or individuals.

🔥 Things You Didn’t Know You Can Do With AI

AI-Powered Meeting Prep (inspired by Zapier)

Zapier’s Executive Assistant shared a workflow that turns AI into your meeting co-pilot. Instead of scrambling through docs, chat threads, and notes before a call, you can:

1. Feed your agenda, shared materials, and chat history into an AI assistant.

2. Ask it to generate a concise (and prioritized) meeting brief.

3. Have it draft key talking points, questions, and decisions.

4. Let it simulate pushback or gaps you might hear.

5. Produce a ready-to-send summary and follow-up plan.

This approach removes friction from prep and lets teams focus on clarity, not context wrangling — especially useful for cross-functional meetings or onboarding new collaborators.

(See the workflow in action here)