Gstack Agents: Open-Source AI Expert Panel Joins Your Google Meet Video Calls

When AI Role-Playing Enters the Real Meeting Room

Imagine this scenario: you're demoing your product in Google Meet, and on the other side of the screen aren't colleagues, but a group of AI experts sporting 3D avatars and offering live critiques — one playing CEO, another acting as Chief Strategy Officer, and yet another simulating a YC partner's "office hours" experience. They take turns speaking, critique your shared screen content in real time, and drop key points into the chat. This is exactly the scenario that Gstack Agents, a recently featured open-source project on Product Hunt, aims to deliver.

The project was built by Anand Balakrishnan, inspired by the "gstack specialists" persona library open-sourced by YC head Garry Tan. It has already received 118 upvotes on Product Hunt, ranking 11th on that day's leaderboard, categorized under Open Source Tools, Developer Tools, and Artificial Intelligence.

Some background on YC Office Hours is worth elaborating: Y Combinator is one of the world's most influential startup accelerators, having incubated well-known companies like Airbnb, Stripe, and Dropbox. One of its core operating mechanisms is "Office Hours" — regular one-on-one meetings between YC partners and portfolio founders, where partners deliver extremely direct, high-density interrogative feedback on product direction, business models, and growth strategies. This format is regarded by Silicon Valley founders as an invaluable "mental stress test." As YC's current CEO, Garry Tan's open-sourced persona library is an attempt to digitally replicate this high-quality feedback model.

What Is Gstack Agents

An AI Expert Team That Can "Join" Your Meeting

Gstack Agents' core selling point is bringing AI out of chat windows and into real video conferences. The expert roster includes 18+ AI personas — CEO, CSO (Chief Strategy Officer), QA Lead, a simulated YC office hours partner, and more. These personas join your Google Meet as voice bots, equipped with 3D virtual avatars.

Their behavioral logic is designed to be quite realistic: they introduce themselves in-persona, take turns speaking, verbally critique the screen content you're sharing, and leave written notes in the meeting chat. In other words, it simulates a real "multi-persona review session" rather than a single AI assistant's Q&A.

Free, MIT Open Source, with Inference Core Under Your Control

This project has several noteworthy technical positioning aspects. First, it's completely free and open-sourced under the MIT license. The MIT license is one of the most permissive licenses in the open-source software world, allowing anyone to freely use, copy, modify, merge, publish, distribute, sublicense, and even sell copies of the software — the only requirement being to include the copyright notice in all copies. Compared to "copyleft" licenses like GPL (which require derivative works to also be open-sourced), the MIT license places virtually no restrictions on commercial use. This means companies can build closed-source commercial products based on Gstack Agents without disclosing their code, significantly reducing legal risks for commercialization and making it easier to attract enterprise developers to the ecosystem.

Second, the team specifically emphasizes that "the brain is your own coding-agent session" — meaning the inference core driving these AI personas isn't locked into a closed service but connects to the user's own coding agent session. Specifically, a "coding-agent session" refers to an AI agent runtime environment centered on code generation and execution capabilities, such as the persistent conversational contexts provided by tools like Cursor, Windsurf, or Claude Code. Under this architecture, Gstack itself only handles the "presentation layer" work — persona orchestration, speech synthesis, meeting integration — while the actual cognitive reasoning is handled by external models. This "thin client" design avoids burdening the project maintainers with expensive GPU inference costs while letting users flexibly choose GPT-4, Claude, Gemini, or other models to power the personas.

The entire system is built on AgentCall, the underlying framework supporting multiple voice agents collaborating in real-time meetings. The core technical challenges AgentCall must solve include: multi-agent turn-taking — ensuring multiple AIs don't talk simultaneously or remain silent too long; low-latency pipelines for real-time text-to-speech (TTS) and speech-to-text (STT); and meeting protocol adaptation — connecting AI agents as compliant participants into Google Meet's WebRTC audio/video streams. Traditional voice AI applications are mostly single-agent, while multi-agent scenarios require solving additional complexities like speaking priority, context sharing, and conversation state synchronization, making AgentCall a critical "orchestration layer" in the tech stack.

Why This Approach Deserves Attention

From "Conversational AI" to "Present AI"

Over the past two years, the interaction paradigm of most AI applications has remained stuck in "you ask, I answer" chat boxes. Gstack Agents represents a different direction: letting AI exist as a "meeting participant" within humans' existing collaboration scenarios. The difference in this "sense of presence" isn't merely formal innovation — when AI can observe your shared screen, interject at appropriate moments, and express opinions from a specific persona's standpoint, it more closely resembles a real review consultant rather than a tool.

From a broader perspective, the evolution of AI interaction paradigms has roughly gone through three stages: command-line style (users input precise commands to get results), conversational (multi-turn natural language interaction, like ChatGPT), and the emerging "present" or "ambient" AI. The core characteristic of present AI is that it no longer waits for users to initiate conversation but continuously perceives environmental context (such as screen content, meeting progress, code changes) and proactively intervenes at appropriate moments. The theoretical foundation of this paradigm traces back to Mark Weiser's 1991 vision of "Ubiquitous Computing" — technology should blend into the environment rather than requiring users to deliberately interact with it. Gstack Agents concretizes this concept into the form of "AI meeting participants," representing an identity leap from tool to colleague.

For founders, this design is particularly appealing. Scenarios like YC office hours, simulated investor interrogation, and product reviews inherently require "multiple different perspectives present simultaneously." Using AI personas to batch-replicate these perspectives costs far less than actually meeting with multiple experts.

The Multi-Perspective Review Value of Role Division

A single large model's advice tends to be a "summed and averaged" middle-of-the-road answer. By having CEO, CSO, QA Lead, and other personas speak separately, Gstack effectively uses prompt engineering and persona configuration to force the model to output opinions from different stakeholder perspectives.

Understanding the technical principles of prompt engineering in role-playing is important here: through carefully designed "system prompts," each persona can be injected with personality settings, professional background descriptions, and behavioral constraint rules. For example, a "CEO persona's" system prompt might include instructions like: focus on revenue models and market size, ask questions from a decision-maker's perspective, prefer concise and direct communication style. Research shows that persona-based prompts can effectively activate domain-specific knowledge patterns the model learned during pre-training, producing more focused and opinionated outputs than generic prompts.

The CEO might focus on business models, the CSO examines strategic moats, and the QA Lead watches for product defects. This structured multi-perspective review often has more practical value than a single generic piece of feedback. However, it's worth noting that persona consistency remains a known weakness of current large language models — phenomena like personas "drifting" or confusing different role positions during long conversations are not uncommon, posing a potential challenge to Gstack's actual experience quality.

Real-World Opportunities and Concerns

Engineering Challenges of Voice Coordination and Real-Time Performance

Stuffing multiple AI voice agents into a real-time video conference is no trivial technical feat. Turn-taking scheduling, voice naturalness, 3D avatar rendering, real-time understanding of shared screen content — every link tests the engineering implementation. The real-time screen understanding component in particular requires continuous parsing of screen frames from the video stream through multimodal models (like GPT-4V or Gemini Pro Vision), with results injected in real time into each persona's conversation context, placing extremely high demands on both latency and computational resources.

The project relies on the AgentCall framework to handle these complexities, but judging from the early-stage buzz of only 10 comments on Product Hunt, it currently looks more like a proof of concept and developer toy, still some distance from a production-grade stable experience.

Feedback Quality Depends on the Model You Connect

A point worth repeatedly emphasizing: since Gstack delegates the inference core to the user's own coding-agent session, the quality of AI feedback is highly dependent on the capability of the connected model and the precision of persona prompt design. This is both its freedom — you can power it with any powerful model — and its source of uncertainty. Done well, persona role-playing yields a high-quality review session; done poorly, it may devolve into several AI avatars talking past each other.

From a practical standpoint, connecting different tiers of models produces significantly different experiences: using GPT-4-level models might generate high-quality strategic insights, while smaller open-source models may fall short in persona maintenance and professional depth. Furthermore, when multiple personas share the same model backend, how to ensure they truly produce differentiated rather than homogenized viewpoints is also a prompt engineering challenge requiring careful tuning.

Conclusion

The significance of Gstack Agents may not lie in how polished it is at this stage, but in how it demonstrates a new form of AI application: letting AI embed itself in humans' real collaboration workflows in a persona-based, voice-enabled, present manner. For those of us accustomed to chat boxes, this is a direction worth watching. Its positioning as MIT open-source, free, and connectable to your own models also lowers the barrier for developers to experiment. If you're looking for a more immersive AI product review experience, this project is at least worth cloning and running once.