Genspark’s Vibe Working Powers Weekly AI Product Releases

In a break from traditional release cycles, startup Genspark has mastered what it dubs “vibe working,” an AI-native approach that lets its 20-person team ship new products and features almost every week. By marrying AI agents with lean team structures, the company claims it could become the fastest-growing startup ever by ARR.

AI-Native Vibe Working for Rapid Innovation

Co-founder and CTO Kaihua Zhu explains that under this model, “everybody is the manager.” Each team member works with a suite of AI agents as reportees, designating tasks end-to-end. They tap into automated code generation and review pipelines to maintain quality while slashing development time.

Super Agent: Next-Gen AI Search

Born as an AI search startup, Genspark quickly pivoted to Super Agent, a platform that picks optimal sub-agents for each task. Powered by Anthropic’s Claude and a mix of other LLMs, Super Agent can complete hours of office work—research, slide decks, calls—in minutes.

• April 11: $10M ARR, 9 days post-launch
• May 2: $22M ARR at one month
• May 19: $36M ARR milestone
• Feature rollouts every week: AI Slides, Sheets, Browser, Design Studio, AI Pods, multi-agent orchestration

Driving Competition and Community Engagement

To prove its edge, Genspark launched a $1M Side-by-side AI Showdown. Users compare Super Agent against ChatGPT Agent and others on identical prompts. Winners earn cash, and the community uncovers edge cases—fueling both product refinement and brand buzz.

Under the Hood: A Multi-Agent Engine

Super Agent uses a mixture-of-experts system across nine LLMs and over 80 specialized tools—from Python code generation to autonomous dialing. An aggregator model then sifts outputs for cost, speed, and accuracy, trimming hallucinations and latency.

With more than 80% of its code AI-generated, Genspark’s team operates like a squad of superheroes. Rigorous code reviews preserve quality, while transparent workflows eliminate politics and friction.

Implications for Enterprise AI Teams

Genspark’s model signals a shift from hierarchical silos to AI-powered squads. Enterprises aiming to replicate its speed must rethink team design, governance, and tooling—embracing AI agents to amplify human creativity and output.

As AI adoption accelerates, leaders need strategic guidance to balance innovation with cost and compliance. QuarkyByte’s analysis can help organizations architect efficient multi-agent pipelines, optimize inference costs, and build agile teams that deliver weekly breakthroughs.