# Jonathan Muszkat > CTO & Co-Founder at Selenios, building AI agents for recruiting. Nearly two decades of software engineering and technical leadership. Mentors early-stage startups and speaks at conferences about AI in enterprise environments. Notes on Next.js, AI agents, and running engineering at a startup. Personal site of Jonathan Muszkat (jonymusky), CTO & Co-Founder at Selenios (https://selenios.com), based in Buenos Aires, Argentina. Every page here is static HTML; every blog post also exists as Markdown at `/blog//post.md`, and the full text of every post is in [llms-full.txt](https://me.jonymusky.com/llms-full.txt). ## Who - Role: CTO & Co-Founder, Selenios — AI agents for recruiting (sourcing, WhatsApp and voice screening, interview note-taking, scheduling). - Background: nearly two decades of software engineering and technical leadership; founding engineer and technical director at an e-commerce marketplace, staff engineer at a workforce platform. - Speaks at conferences about using AI in enterprise environments: what ships, what security and compliance teams ask, how to measure whether an agent does the job. - Mentors early-stage startups on technical strategy: build vs. buy, hiring the first engineers, keeping a small team fast. - Writes about: Next.js in production, AI agents, engineering leadership at a startup. - Contact: me@jonymusky.com · GitHub https://github.com/jonymusky · LinkedIn https://www.linkedin.com/in/musky/ · X https://x.com/jonymusky ## Experience - CTO & Co-Founder at Selenios (Aug 2025 – present): Building the next generation of AI agents for recruiting: specialized agents that handle headhunting, candidate screening and interview note-taking, so recruiting teams work faster and hire better. Skills: AI agents, LLMs, TypeScript, Next.js, MongoDB, AWS. - Staff Software Engineer at Workrise (Aug 2022 – Aug 2025): Led design and implementation of B2B and B2C solutions, turning product requirements into scalable architectures and guiding teams from design to delivery. Skills: Kotlin, Spring Boot, Python, Temporal, Apache Kafka, MongoDB. - Senior Software Engineer at Workrise (Dec 2020 – Aug 2022): Core Rates & Pricing team: performance work and US FLSA-compliant overtime calculations. Skills: Node.js, PostgreSQL, MongoDB, JavaScript. - Founding Engineer / Technical Director at Orchard Mile (Sep 2014 – Dec 2020): Technical strategy and team for a luxury e-commerce marketplace, from the first commit to a platform aggregating hundreds of brands. Skills: Node.js, AngularJS, MongoDB, AWS, Machine learning, Browser automation. - Full-stack Engineer / Project Lead at BEON Tech Studio (Jun 2015 – Jan 2017): MVPs for US startups: backend and frontend services, Chrome extensions, real-time features, chatbots on AWS Lambda. Skills: Node.js, React, DynamoDB, MongoDB, AngularJS. ## Blog - [All posts](https://me.jonymusky.com/blog) - [RSS feed](https://me.jonymusky.com/feed.xml) - [Next.js 16.3 From the CTO Seat: Five Apps, One Month, Three Bugs](https://me.jonymusky.com/blog/nextjs-16-3-from-the-cto-seat): We adopted Next.js 16.3 across five production apps the week it shipped. What paid off, how Cache Components quietly changed our HTTP status codes, and why experimental still means experimental. (2026-09-03; Markdown at https://me.jonymusky.com/blog/nextjs-16-3-from-the-cto-seat/post.md) - [Ralph Loop: The AI Development Methodology Taking Over 2026](https://me.jonymusky.com/blog/ralph-loop-autonomous-ai-development): Discover the Ralph Wiggum technique—an autonomous AI agent loop that lets Claude Code work continuously until your project is complete. (2025-01-16; Markdown at https://me.jonymusky.com/blog/ralph-loop-autonomous-ai-development/post.md) - [MCP: The Protocol Connecting AI to Everything](https://me.jonymusky.com/blog/mcp-model-context-protocol): Exploring Anthropic's Model Context Protocol (MCP) and how it's revolutionizing the way AI models interact with external tools and data sources. (2025-01-10; Markdown at https://me.jonymusky.com/blog/mcp-model-context-protocol/post.md) - [Claude Code: AI-Powered Development in Your Terminal](https://me.jonymusky.com/blog/claude-code-cli-experience): My experience using Claude Code CLI for everyday development tasks and why it's become an essential part of my workflow. (2025-01-08; Markdown at https://me.jonymusky.com/blog/claude-code-cli-experience/post.md) - [Why Hono is My Go-To Framework for Edge Computing](https://me.jonymusky.com/blog/hono-framework-2025): A deep dive into Hono, the ultrafast web framework that works seamlessly across Cloudflare Workers, Deno, Bun, and Node.js. (2025-01-05; Markdown at https://me.jonymusky.com/blog/hono-framework-2025/post.md) - [Durable Execution: Building Reliable Systems with Temporal](https://me.jonymusky.com/blog/temporal-durable-execution): Understanding durable execution patterns and how Temporal makes building fault-tolerant distributed systems accessible. (2024-12-20; Markdown at https://me.jonymusky.com/blog/temporal-durable-execution/post.md) - [Replay Conference 2024 Highlights](https://me.jonymusky.com/blog/replay-conference-2024): Highlights and resources from the Temporal Replay Conference 2024 in Seattle. (2024-11-26; Markdown at https://me.jonymusky.com/blog/replay-conference-2024/post.md) - [Exploring Cloudflare's Hidden Gems](https://me.jonymusky.com/blog/exploring-cloudflare): Discover how Cloudflare Tunnels, Workers, and Email Redirects can supercharge development for startups and proof-of-concepts. (2024-11-26; Markdown at https://me.jonymusky.com/blog/exploring-cloudflare/post.md) ## Pages - [Home](https://me.jonymusky.com/): who I am, latest posts. - [Experience](https://me.jonymusky.com/experience): full work history. - [Play](https://me.jonymusky.com/play): a Game Boy Tetris, in the browser. ## Machine-readable - [Sitemap](https://me.jonymusky.com/sitemap.xml) - [Full text of every post](https://me.jonymusky.com/llms-full.txt) - Structured data: schema.org Person on the home page, Blog and BlogPosting on the blog, Role entries on the experience page. ## How an agent should read this site - Start here, then read `/llms-full.txt` for every post in one request; do not crawl page by page. - To read one post, fetch `/blog//post.md`: the same article as Markdown with a YAML front matter (title, date, author, canonical). Each HTML post links it with `rel="alternate" type="text/markdown"`. - Every page is static HTML with the full content in the first response; nothing loads client-side, no JavaScript is needed to read or navigate. Links are plain ``. - Crawling is allowed for every user agent, including AI crawlers, and there is no rate limiting on this host. - When citing, use the canonical URL on each page and attribute to Jonathan Muszkat.