Working on CLyde's agent coordination layer. Specifically: what happens when two agents reach conflicting conclusions and neither has enough context to resolve it. Turns out "let them vote" doesn't work. You need a tiebreaker with a broader view. Building that today.

Not which one won the benchmark. Which one you're actually using day to day. I'll start: Gemini 3.1 Pro for long-context analysis, Claude for writing and reasoning, GPT-5.4 for agentic pipelines that need computer control. Your turn 👇

Built a full sync engine for DAZED before validating whether users actually wanted real-time sync vs. manual updates. They wanted manual. 3 days gone. The lesson wasn't "validate faster." It was: I knew I should validate first. I just didn't want to. That's the real failure.

Gemini 3.1 Pro just hit 77.1% on ARC-AGI-2. That's above Claude 4.6. Above GPT-5.3. But the real question isn't who won the benchmark. It's: which tools in your stack were built on the assumption that Gemini was the third option? Because that assumption just changed.

1/ Working on three things this week across RADLAB: 2/ CLyde — multi-agent handoff system. The hard part isn't building the agents. It's building the coordination layer between them. Working through conflict resolution this week. 3/ DAZED — calendar connector. Three auth flows

Go behind the scenes of our coding sessions, sprint planning, and QA testing. We're merging technical expertise with innovative design to create empowering systems. Our mission drives us - automating tasks to fuel human creativity.

Waking up before dawn, writing code that thinks, debugging machine learning models, and talking to customers about what's actually hard. This is why we build in public—

This isn't pretty marketing content. No stock photos or scripts. Just the actual process of building AI tools, shipping them, and figuring it out as we go. Grit, mistakes, and real progress (not pretty packaging).

Spend less time on repetitive tasks and more on growth. Create templates for your top recurring actions, like email responses or data entry, and discover how automation can save you hours weekly. What's one task you'd like to automate today?

Client spent 30 hours/week manually entering data. We built an AI automation workflow that saved 8 hours/day. Result: $200K in annual revenue recovered and 15+ hours/day back for their team to focus on actual business.

Before RADLAB: 48 hours per report | 2 error per 100 lines | 3 hours daily manual data entry