layout: post
title: Transform Your Startup with Y Combinator's AI-Powered Guidance
subtitle: Leverage Gest in Claude to Supercharge Your SaaS Development
cover-img: /assets/img/crag_chatbot_local_spc_agent_cover1.png
thumbnail-img: /assets/img/yc_y_combinator_sas_office_hour_advice.png
share-img: /assets/img/yc_y_combinator_sas_office_hour_advice.png
tags: [AI, Startups, SaaS, Y Combinator, Claude, Development, Productivity, Entrepreneurship]
author: Kayhan Kaptan - Medical Physics, Quality Control, Data Science and Automation

Have you ever dreamt of getting into Y Combinator (YC), the legendary startup accelerator that has nurtured giants like Airbnb, Stripe, OpenAI, and Coinbase? Founded in 2005 by Paul Graham, YC boasts over 4,000 successful startups and a collective valuation exceeding $600 billion. The wealth of experience and advice accumulated over two decades is invaluable for any aspiring entrepreneur.
Imagine having access to this compressed wisdom directly in your development environment. Thanks to Gary Tan, the CEO of YC, this is now a reality. He has open-sourced an incredible repository called Gest, a collection of “skills” and “power tools” designed to bring YC’s 20 years of startup expertise into your terminal. With over 87,000 stars, Gest is a game-changer for SaaS developers.
This blog post will guide you through how Gest works and demonstrate its power by putting a real SaaS project through its paces, specifically using the crucial “Office Hour” feature.
What is Gest?
Gest is more than just a collection of files; it’s a comprehensive framework embodying YC’s philosophy and processes. It includes:
- 23+ Skills: Encapsulating YC’s core methodologies and advice.
- 8 Power Tools: For tasks like creating mockups, generating images, and design (e.g., Nano Banana, DALL-E inspired tools).
- A Sprint Pipeline: A structured workflow covering sync, plan, build, review, test, shape, and reflect phases, all imbued with YC’s operational philosophy.
- Compatibility: Works seamlessly with various AI assistants like Open Claude, Codex, and Cursor.
- Six Virtual Roles: Imagine having a virtual YC engineering team directly in your terminal, with a CEO, Product Manager, Designer, Security expert, QA specialist, and Manager, all ready to scrutinize and guide your project.
The promise is bold: to encode a YC engineering team directly into your terminal.
Step-by-Step Guide to Using Gest
Let’s dive into how to set up and leverage Gest, focusing on the “Office Hour” skill.
Installation
- Clone the Repository: Gest is typically available as a command-line tool. You’ll need to execute a simple command in your terminal to install it. This command usually involves cloning the repository and setting up the environment.
(The video demonstrates a quick installation process, which may involve a single
curlornpmcommand depending on the tool’s distribution method.) - API Keys Configuration: Gest might require API keys for certain functionalities, especially if it integrates with tools like Nano Banana for image or design generation. You’ll be prompted to enter these during installation.
- Verify Installation: Once installed, you should see Gest’s components integrated into your environment, often within specific directories like
skillsin your AI assistant’s configuration (e.g., Claude’s skills directory).
The “Office Hour” Skill: A Deep Dive
The “Office Hour” skill is particularly powerful. It simulates a conversation with a YC partner, challenging your assumptions and guiding you with data-driven insights before you even write a line of code. This is crucial for developers who often jump straight into building without thoroughly validating their hypotheses.
Scenario: Let’s imagine a SaaS project that’s generating €200 MRR but suffering from low conversion rates despite a high number of free users. The initial thought might be to redesign the onboarding process, remove free credits, and force users to a direct payment wall.
Using Gest’s Office Hour:
- Initiate Office Hour: In your development environment (e.g., Claude Code), trigger the
Office Hourskill. This will start an interactive session. - Present Your Problem: Explain your SaaS’s current state and your proposed solution. For instance: “My SaaS costs more than it earns (€200 MRR). I have many free users but low conversion. I want to remove free credits and implement a demo + paywall directly.”
- YC’s Initial Challenge: Gest will immediately challenge your assumptions. It will likely say something like: “You’ve presented a problem and a solution, but we haven’t diagnosed the root cause. If you implement this now, it might work, and you won’t know why, or it might fail, and you’ll have wasted two weeks. Let’s do some diagnostic first.”
- Data-Driven Questions (Q1: Reality Check): Gest will then ask for concrete data, rejecting approximations.
- “How many paying customers do you have per month?”
- “Of these, how many have been paying for more than two months? Do they churn after the first month?”
- “How many actively use the product, not just pay and forget?”
- “Has anyone contacted you when something broke or when features failed?”
- Analyze Responses: After you provide your answers (e.g., “About ten paying users, over 50% retention for more than two months, active usage, and yes, one user even chased me on LinkedIn when credits weren’t applied.”), Gest provides feedback:
- “Solid. The LinkedIn chase is a real signal – someone actively hunting you down for their credits shows real demand. 50% retention on 10 paying users isn’t catastrophic. Your product has real value for a specific segment. Your problem is indeed conversion.”
- Further Questions (Q2: Status Quo): Gest pushes deeper, focusing on user behavior post-free credits.
- “What do these free users do after consuming their free credits?”
- “How many see the pricing page after credits expire?”
- “How many start the checkout process and abandon it?”
- “How many create multiple accounts for more free credits?”
- “How many don’t even consume all their credits?”
- Identify the Real Problem: If you answer “I don’t know” to several questions, Gest will highlight the lack of data as a critical issue. For example, if 40 users abandon checkout but only 10 convert, Gest might tell you:
- “Your initial hypothesis that people don’t see the pricing enough is contradicted by the 40 checkout abandonments. They see it, they click to pay, and then they leave. Forcing them to that page 30 seconds earlier won’t change their decision; it will just add friction.”
- “Your true problem is likely mass leakage at checkout for an unknown reason, not an onboarding issue.”
- “This is a classic pattern: early-stage startups blame onboarding because it’s visible and exciting to redesign, rather than addressing the less ‘sexy’ but more impactful analytics dashboard and session recording.”
- The “Stop” Command: Finally, Gest might issue a “STOP” command, forcing you to prioritize.
- “Imagine I forbid you from touching onboarding for 30 days. What single thing could you fix this week that would have the highest probable impact on MRR, based on what we’ve just discussed?”
- The answer should be clear: “Fix the checkout abandonment issue.” Gest emphasizes that if you hesitate, you haven’t internalized your data (or lack thereof).
This process clearly demonstrates how Gest, acting as a virtual YC partner, guides you away from premature optimization and towards data-driven decisions.
Beyond Office Hour: The Full Pipeline
The Office Hour is just the beginning. Gest offers a complete pipeline:
- Plan (CEO, Design): Strategic planning and design mockups.
- Build: Development phase.
- Review: Code review (can use other models like Codex for this).
- Test: QA, including launching real browser tests (e.g., with Playwright).
- Shape & Reflect: Sprint retrospective.
Crucially, this pipeline is cumulative. You can’t move to step 3 if step 2 isn’t properly validated. If Office Hour tells you to contact users because you haven’t gathered enough data, the plan CEO skill will remind you of that before letting you proceed. This pushes developers, who often shy away from user interaction, to engage with their audience.
The YC Philosophy: “Boil the Ocean”
Gest embodies YC’s “boil the ocean” philosophy, as discussed by Gary Tan. In the age of AI, we can achieve incredible things. We shouldn’t limit ourselves to small ideas but instead leverage all available tools to push our concepts to their fullest potential. This includes using AI for design, image generation, and making the most of every insight available.
Conclusion
If you’re a solo SaaS founder or a developer with startup ambitions, Gest is an indispensable tool. It provides the rigorous, data-driven, and user-centric advice you’d get from Y Combinator, right within your terminal. It challenges assumptions, prevents wasted effort, and guides you toward building a truly valuable product.
🔍 Discover Kaptan Data Solutions — your partner for medical-physics data science & QA!
We're a French startup dedicated to building innovative web applications for medical physics, and quality assurance (QA).
Our mission: provide hospitals, cancer centers and dosimetry labs with powerful, intuitive and compliant tools that streamline beam-data acquisition, analysis and reporting.
🌐 Explore all our medical-physics services and tech updates
💻 Test our ready-to-use QA dashboards online
Our expertise covers:
🔬 Patient-specific dosimetry and image QA (EPID, portal dosimetry)
📈 Statistical Process Control (SPC) & anomaly detection for beam data
🤖 Automated QA workflows with n8n + AI agents (predictive maintenance)
📑 DICOM-RT / HL7 compliant reporting and audit trails
Leveraging advanced Python analytics and n8n orchestration, we help physicists automate routine QA, detect drifts early and generate regulatory-ready PDFs in one click.
Ready to boost treatment quality and uptime? Let’s discuss your linac challenges and design a tailor-made solution!
Get in touch to discuss your specific requirements and discover how our tailor-made solutions can help you unlock the value of your data, make informed decisions, and boost operational performance!
Comments