Subhansh Kumar@near

Friday is built to solve a simple but important problem: most AI tools are fragmented, forgetful, and reactive. They can answer questions, but they don’t reliably remember context, coordinate complex workflows, or carry tasks through to completion. Friday brings those capabilities together in one system designed to feel more like a working intelligence than a chatbot. A core part of Friday is its memory architecture. Users can feed information into its core directory, and Friday retains that context across sessions, so the more you use it, the more useful it becomes. This makes it better for long-running projects, repeated workflows, and personalized assistance. Friday also uses multi-agent orchestration to handle complex work. It can coordinate 30+ specialized agents at once for coding, research, automation, system control, debugging, and verification. Instead of stopping at a response, Friday decides which agent to invoke, how to chain them together, and how to continue until the task is actually done. The architecture is inspired by neuroscience and includes ideas like memory layers, reasoning loops, proactive behavior, reflection, and long-chain task execution. Friday is also designed to support voice interaction and a holo builder for gesture-based 3D interaction, which explores a more natural future for human-AI collaboration. Overall, Friday is my attempt to build an AI platform that remembers, reasons, coordinates, and executes — not just one that talks.

Friday is an autonomous AI system built to go beyond simple chat-based assistants. Most AI tools today are reactive, forgetful, and fragmented across different apps and workflows. Friday solves this by combining persistent memory, multi-agent orchestration, reasoning, voice interaction, tool use, and long-term context into one unified system. A core feature of Friday is its memory architecture. Users can feed information into Friday’s core directory, and it remembers that context across sessions. This makes Friday more useful over time because it can retain preferences, project context, and learned information instead of starting from scratch every time. Friday also uses multi-agent orchestration to handle complex tasks. It can coordinate 30+ specialized agents at once for coding, research, automation, system tasks, and verification. Rather than acting like a simple chatbot, Friday decides which agent to invoke, how to chain them together, when to verify outputs, and how to continue until the task is completed. The system is inspired by neuroscience and built around ideas like memory layers, reasoning loops, active inference, reflection, and proactive behavior. Friday is designed not just to respond to prompts, but to think through tasks, anticipate needs, and improve from repeated usage. Friday also includes a holo builder for gesture-based 3D interaction, giving it a more spatial and experimental interface layer. This explores what the future of human-AI interaction could look like beyond text boxes. Overall, Friday is my attempt to create an AI system that feels more like a working intelligence than a chatbot: one that remembers, coordinates, executes, and learns.

Friday is a next-generation autonomous cognitive AI assistant built by a self-taught 17-year-old developer with no university education. Built on AMD Developer Cloud and powered by Gemini 2.5 Flash Live API, Friday runs locally and delivers real-time voice interaction with a cognitive cycle of perception, memory, reasoning, execution, reflection, and learning. Friday is more than a chatbot or API wrapper — it is a cognitive architecture inspired by neuroscience and AI research. It includes memory, planning, reasoning, and self-improvement, with modules inspired by the Free Energy Principle, hierarchical active inference, world models, analogical reasoning, causal reasoning, and cognitive architecture principles. Its memory architecture combines neural, episodic, vector, procedural, working, and global workspace memory. Memories persist across sessions, support semantic recall, and are consolidated through offline replay and learning. The cognitive coding engine analyzes codebases, builds semantic graphs, plans with expected free energy, simulates execution before running, and performs structured debugging with root-cause reasoning. Friday also includes a multi-layered cybersecurity pipeline with static analysis, adversarial verification, data-flow tracing, and exploit-class reasoning, all gated by explicit user confirmation and restricted from unauthorized or local targets. Beyond cognition and security, Friday supports gesture-controlled 3D holographic interaction, globe navigation, music control, and emotionally adaptive voice interaction. Friday demonstrates what a single self-taught builder can create with the right infrastructure: a locally running, benchmarked, continuously improving cognitive AI system built on AMD Developer Cloud. Built by one developer. 17 years old. No university. No team. 18 days.

🌍 Hybrid Hackathon | 💻 Online Build: May 25–30,2026 | 📍 SF Onsite Finale: May 30–31,2026 We’ve unlocked the web for your AI. Now build what wasn’t possible before. Build AI agents with live web data, then join the SF onsite finale at Bright Data’s new Web Data Loft. 🚀 The onsite experience will take place during the opening of Bright Data’s new SF space, The Web Data Loft. 📍 On-site Venue (May 30–31): The Web Data Loft 625 2nd St, San Francisco, CA 🏆 $18,300 cash prize + Bright Data AI Startup Program opportunities 🍕 Free food, drinks, networking, workshops, mentorship, and live demos 🎁 Onsite attendees are automatically entered into a live giveaway for 4 × $100 Amazon gift cards. 🧑‍💻 Apply now to build AI agents powered by real-time web access.

Build solutions that improve how software is built. Work with an AI that understands your codebase, helping you reduce repetitive work and build with real context. ⏳ Build your project in 48 hours. 🚀 Get hands-on with IBM Bob and explore how AI fits into real development workflows before it becomes standard. 💡 Build tools and workflows that developers would actually use. ⭐ Learn directly from experts throughout the event. 🤝 Join a team or build solo. 💰 Prize pool: $10,000 🧑🏻‍💻 Register before the Kick-Off Stream and start building with Bob.

⏱️ Build the next generation of Autonomous Agents in the heart of Europe’s AI Revolution. 📅 May 13 – 20, 2026 May 13 – 19: Online Build Phase. May 19: An exclusive on-site build day at Milan AI Week for selected participants. May 20: Awards Ceremony. 📍On-site Venue (May 19–20): Fiera Milano (Rho), Milan, Italy 🤝 Join solo or with a team. 🎟 FREE Milan AI Week conference ticket included for participants. 💰 $32,000+ prize pool 🧑‍💻 Apply now to turn your autonomous agent into a market-ready enterprise solution.