--- title: "Sezer Yavuz — Comprehensive site content" description: "Software engineer in San Francisco. Founder of ModulexAI, the AI workspace that comes pre-connected to your tools. Previously co-founded ThinkBuddy (Product Hunt #1)." url: "https://sezeryavuz.com" type: "comprehensive" last_updated: "2026-06-30T08:42:11-05:00" --- # Sezer Yavuz **AI Engineer & Software Engineer · San Francisco, CA** _Last updated: June 30, 2026._ > Software engineer in San Francisco. Founder of ModulexAI, the AI workspace that comes pre-connected to your tools. Previously co-founded ThinkBuddy (Product Hunt #1). This document contains the full content of https://sezeryavuz.com. Every page on the site (home, about, work, now, writing) is included below. The site itself is at https://sezeryavuz.com; the canonical Markdown URL is https://sezeryavuz.com/sezeryavuz.md. --- ## Table of contents - [About](#about) - [Quick facts](#quick-facts) - [Now](#now) - [Selected work](#selected-work) - [Recognition](#recognition) - [Education](#education) - [Frequently asked](#frequently-asked) - [Writing](#writing) - [Contact](#contact) - [Recent changes](#recent-changes) --- ## About Sezer Yavuz is a software engineer based in San Francisco. He is the founder of ModulexAI, an AI workspace that comes pre-connected to a team's tools, and previously co-founded ThinkBuddy, an AI productivity platform that reached Product Hunt #1. ### What I do I am a software engineer in San Francisco focused on AI systems. My current company, ModulexAI, is an AI workspace that comes pre-connected to a team's tools: a chat assistant that answers and acts on their data, plus a visual workflow builder with an AI copilot that builds the graph from plain language. The bet is that the bottleneck in AI automation is connection and setup, not the model, so we made connectivity the product. I work across distributed systems, the runtime that executes it all, multi-agent orchestration, and the developer experience around all three. ### How I got here I grew up in Turkey and studied Industrial Engineering at Istanbul Okan University, where I completed the program on a full merit scholarship with a 3.7 GPA and wrote my thesis on rule-based fuzzy systems for anthropometric data. Software was always the actual interest. Early roles at Creative Dock (a Prague-based venture builder) and ERC Barkod (industrial IoT) were where I learned to ship: building React frontends for a banking app, then real-time inventory algorithms over RFID and IoT sensor streams that cut processing time by 40%. ### ThinkBuddy In 2023, I co-founded ThinkBuddy. We built an AI productivity platform that unified 15+ leading AI models behind a voice-first interface. Over two years we grew it from concept to 20,000+ active users and 4,000+ paying subscribers. ThinkBuddy hit #1 on Product Hunt on February 27, 2024. The technical core was a RAG pipeline with an MCP server enabling context-aware multi-source AI responses, and a document intelligence layer that handled OCR, file parsing, and LLM-based extraction across many file formats. ### ModulexAI I started ModulexAI in October 2025. The thesis: most teams stall before AI does any real work, not on the model, but on the setup. Connecting each tool means an account, an API key, and glue code, and one credential per tool is never enough. So I made connectivity the product. ModuleX is an AI workspace that ships pre-connected to 200+ integrations, with ModuleX Key supplying managed credentials for premium tools (billed as credits) and many accounts of the same tool usable at once, so you can run several Gmail accounts through one assistant. It is one engine behind two surfaces: a chat assistant that knows your data and acts on it, and a visual workflow builder where an AI copilot, Composer, builds the graph from plain language. Under the hood it is a runtime that compiles those definitions to executable LangGraph statecharts; the same workflow runs in chat, on the canvas, or as a REST API with TypeScript and Python SDKs, built once and run anywhere. ## Quick facts - **Name:** Sezer Yavuz (legal: Sezer Ufuk Yavuz) - **Role:** AI Engineer & Software Engineer - **Location:** San Francisco, CA - **Currently:** Founder of [ModulexAI](https://modulex.dev) - **Previously:** Co-Founder at [ThinkBuddy](https://thinkbuddy.ai) - **Education:** BSc Industrial Engineering, Istanbul Okan University (2017 – 2022) - **Email:** s.ufukyavuz@gmail.com --- ## Now _What Sezer is focused on this month (as of June 2026)._ Building ModulexAI. Most of the day is the connectivity layer: widening the pre-connected integration catalog and making ModuleX Key, the managed-credential system for premium tools, painless. No signups, no keys to hunt, billed cleanly through credits. On the harder-problems side: Composer, the AI copilot that builds and edits workflow graphs from plain language. The interesting unsolved part is letting it mutate live workflows safely, applying graph diffs without breaking running checkpoints. There is a clean answer hiding in there somewhere. Outside of code: working with the first integration partners on the TypeScript and Python SDKs, and writing more often. If any of this is useful to you or your team, the email at the bottom of the page works. Source: https://sezeryavuz.com/now --- ## Selected work ### ModulexAI **The AI workspace that comes pre-connected to your tools.** - Role: Founder & Lead Engineer - Period: Oct 2025 – Present - Status: Active - Live: https://modulex.dev - Case study: https://sezeryavuz.com/work/modulexai **Context.** Most teams stall before AI does any real work, not on the model, but on the setup: an account and an API key for every tool, glue code to wire them, and one credential per tool that's never enough. ModulexAI makes connectivity the product. It ships pre-connected to 200+ integrations, brings managed credentials for premium tools, and runs many accounts of the same tool at once, exposed as a chat assistant that acts, a visual workflow canvas, and an API, all on one engine. **What I built:** - Batteries-included connectivity: 200+ pre-wired integrations available to every assistant and workflow out of the box, with OAuth 2.0 flows so teams skip the per-tool account-and-key setup entirely. - ModuleX Key, a managed-credential layer that brokers access to premium tools (Tavily, Firecrawl, ElevenLabs, and a list growing weekly) through a metered credit system, so users get the premium stack with no signups or keys to hunt. - Multi-credential support: many accounts of the same integration, usable simultaneously inside one assistant or workflow, for example several Gmail accounts through a single assistant. - Runtime compiler that translates workflow definitions into executable LangGraph statecharts, with dynamic TypedDict state generation, a `{{node.path}}` DSL for cross-node reference resolution, and parallel FOREACH fan-out via the Send API. - Composer, a multi-agent AI copilot (DeepAgents) that plans, generates, and mutates the workflow graph from natural language on a visual canvas (xyflow / React Flow), streamed via SSE over Redis Pub/Sub. - Execution layer supporting 12+ node types (LLM, RAG, Conditional, Loop, Guardrails, Interrupt) with per-node retry, credit tracking, and PostgreSQL-backed checkpointing; hybrid keyword and semantic RAG over pgvector grounds assistants in a team's data. - Distributed real-time collaboration engine (Socket.io + Redis Pub/Sub) with node-level locking and JSON-patch state sync (fast-json-patch), so multiple people edit one graph live. - Credential vault with per-credential encryption via HKDF-SHA256 key derivation; one engine exposes every workflow as a chat assistant, a canvas, and a callable REST endpoint, with zero-dependency TypeScript and Python SDKs for embedding ModuleX in third-party products. **Outcome:** - Stage: Live in production - Recognition: Product Hunt #6 Product of the Day (2026) - Pre-connected integrations: 200+ - Surfaces, one engine: Chat · Canvas · API ### ThinkBuddy **An AI productivity platform with voice-first interaction.** - Role: Technical Co-Founder - Period: Feb 2023 – Oct 2025 - Status: Previous - Live: https://thinkbuddy.ai - Case study: https://sezeryavuz.com/work/thinkbuddy **Context.** ThinkBuddy started as a way to make the best AI models accessible without juggling tabs and accounts. We unified 15+ leading models behind a single voice-first interface, then layered context-aware retrieval, document understanding, and multi-source responses on top. **What I built:** - Unified interface integrating 15+ leading AI models with voice-first interaction capabilities. - RAG pipeline with an MCP server enabling context-aware, multi-source AI responses across user sessions. - Document intelligence pipeline with OCR, file parsing, and LLM-based content extraction across multiple file formats. - Subscription billing, account management, and a usage layer that handled fair-use and pro-tier accounting across model providers with very different pricing models. **Outcome:** - Active users: 20,000+ - Paying subscribers: 4,000+ - Recognition: Product Hunt #1 Product of the Day (Feb 27, 2024) - AI models integrated: 15+ Source: https://sezeryavuz.com/work --- ## Recognition - **2026** — Product Hunt #6 Product of the Day · ModulexAI - **2024** — Product Hunt #1 Product of the Day · ThinkBuddy - **2023** — Generative AI Hackathon — 2nd place · Newsea - **2021** — Tilburg University Challenge — 3rd of 80 · ISOLA ## Education **BSc Industrial Engineering** — Istanbul Okan University (2017 – 2022) ## Work history ### Founder & Lead Engineer — ModulexAI (2025 – Present) _San Francisco_ Building the AI workspace that comes pre-connected to 200+ tools: a chat assistant that answers and acts on a team's data, plus a visual workflow builder with an AI copilot that builds the graph from plain language. The thesis is that the bottleneck in AI automation is connection and setup, not the model. ### Technical Co-Founder — ThinkBuddy (2023 – 2025) _San Francisco_ Built and scaled an AI productivity platform from concept to 20,000+ users and 4,000+ paying subscribers, with a #1 Product of the Day on Product Hunt. Unified interface across 15+ AI models with voice-first interaction. ### Software Engineer — ERC Barkod (2021 – 2023) _Istanbul_ Built real-time inventory tracking over RFID and IoT sensor streams, plus the analytics pipeline aggregating live operational metrics across multi-site ERP deployments. ### Junior Software Engineer — Creative Dock (2021) _Prague_ Built React and TypeScript frontends for a digital banking product at a Prague-based venture builder. --- ## Frequently asked ### Who is Sezer Yavuz? Sezer Yavuz is a software engineer based in San Francisco, California. He is the founder of ModulexAI, an AI workspace that comes pre-connected to a team's tools. Previously, he co-founded ThinkBuddy, an AI productivity platform that scaled to over 20,000 users and reached Product Hunt #1 in February 2024. ### What is ModulexAI? ModulexAI is an AI workspace that comes pre-connected to 200+ integrations. It pairs a chat assistant that answers and acts on your data with a visual workflow builder whose AI copilot, Composer, builds workflows from plain language. ModuleX Key supplies managed credentials for premium tools, and many accounts per integration run at once. ### What was ThinkBuddy? ThinkBuddy was an AI productivity platform that unified 15+ leading AI models behind a voice-first interface. It grew to 20,000+ active users and 4,000+ paying subscribers, and was named Product Hunt's #1 Product of the Day on February 27, 2024. Sezer was a co-founder from February 2023 to October 2025. ### Where is Sezer Yavuz based? Sezer Yavuz is based in San Francisco, California. He works on ModulexAI in a hybrid arrangement between SF and remote. ### What is Sezer Yavuz currently working on? Sezer is building ModulexAI, the AI workspace that comes pre-connected to your tools. The current focus is the ModuleX Key program for managed premium-tool credentials, deepening Composer's natural-language workflow building, and shipping the TypeScript and Python SDKs to early integration partners. ### What is Sezer Yavuz's technical background? Sezer's primary stack is Python, TypeScript, and Rust. His AI engineering work uses LangChain, LangGraph, the Model Context Protocol (MCP), retrieval-augmented generation, and vector search. On infrastructure, he works with PostgreSQL, Redis, pgvector, Docker, Kubernetes, and the major cloud providers. ### How can I contact Sezer Yavuz? The fastest way to reach Sezer is by email at s.ufukyavuz@gmail.com. He is also reachable on X as @sezerufukyavuz, on LinkedIn at linkedin.com/in/ufukyavuzz, and on GitHub at github.com/sezeryavuz. --- ## Writing Notes and essays. Mostly about building agentic AI systems and the patterns I keep returning to. _Nothing published locally yet._ Older long-form writing is on Medium: https://medium.com/@sezerufukyavuz Source: https://sezeryavuz.com/writing --- ## Contact - **Email:** s.ufukyavuz@gmail.com - **GitHub:** https://github.com/sezeryavuz - **X:** https://x.com/sezerufukyavuz - **LinkedIn:** https://linkedin.com/in/ufukyavuzz - **Medium:** https://medium.com/@sezerufukyavuz --- ## Recent changes - `b980db8` June 30, 2026 — Strengthen SEO/GEO: Organization entities, fix favicon + article Markdown - `e927c28` June 30, 2026 — Align site copy with branding dossier; refine ModuleX work card - `e097565` June 30, 2026 — Update sezer-profile - `912c33f` June 30, 2026 — Add project logos to the Work nav dropdown items - `73534d8` June 30, 2026 — Reposition ModuleX as the connected AI workspace across the site - `d82013e` June 30, 2026 — Fix Work nav dropdown and update profile content - `34e4a3d` June 30, 2026 — work panel - `07de4dc` May 14, 2026 — Initialize project infrastructure --- _Source: https://sezeryavuz.com_ _Canonical Markdown: https://sezeryavuz.com/sezeryavuz.md_ _Per-article Markdown: https://sezeryavuz.com/writing/.md_