Is AI killing open-source software? The data makes it clear: open source is thriving. One business model is not.
About us
Knitli is rethinking how humans and AI work together. 𝗣𝗿𝗼𝗯𝗹𝗲𝗺: AI agents are powerful but inefficient. Developers spend hours feeding context, and AI wastes thousands of tokens on tasks it struggles with. 𝗦𝗼𝗹𝘂𝘁𝗶𝗼𝗻 / 𝗞𝗻𝗶𝘁𝗹𝗶’𝘀 𝗮𝗽𝗽𝗿𝗼𝗮𝗰𝗵: We’re building the “context layer” — giving AI exactly what it needs upfront. This reduces wasted effort, lowers costs, and lets AI focus on what it does best. 𝗩𝗶𝘀𝗶𝗼𝗻 (𝗯𝗶𝗴𝗴𝗲𝗿 𝘁𝗵𝗮𝗻 𝗱𝗲𝘃𝘀): AI tools should work for everyone. No prompt engineering. No technical expertise. Just tools that make sense — for AI, for humans, for both. 𝗣𝗵𝗶𝗹𝗼𝘀𝗼𝗽𝗵𝘆: This isn’t an AI problem. It’s a design problem. At Knitli, we design with warmth and empathy. Technology should serve humans, not the other way around. 𝗙𝗼𝘂𝗻𝗱𝗲𝗿 𝗻𝗼𝘁𝗲: Currently in development by Adam Poulemanos — former intelligence executive turned full-stack developer, passionate about making AI truly useful for everyone.
- Website
-
https://knitli.com
External link for Knitli
- Industry
- Software Development
- Company size
- 1 employee
- Headquarters
- Maryland
- Type
- Privately Held
- Founded
- 2025
- Specialties
- AI, software, open source, intelligent context, and empathy-driven design
Locations
-
Primary
Get directions
Maryland, US
Updates
-
Most AI coding tools fail not because of bad models or prompts, but because of poor context management. I call it the "goldfish problem" - your AI forgets what it just learned and has to re-read everything, burning tokens and slowing down with each interaction. At Knitli, I'm tackling this through context engineering. It's the difference between AI tools that actually help versus ones that just rack up API costs. The hard truth: summarization loses critical details. RAG retrieves irrelevant context. Caching helps but isn't a complete solution. Each approach has real tradeoffs, and there's no universal fix yet. What works? Keep sessions focused. Provide explicit context upfront. Match your tools to tasks based on how they handle context windows. Understanding these limitations is the first step to building AI workflows that don't waste your time and money. I built CodeWeaver, Knitli's open-source code search tool, with these principles in mind. It's designed for the Model Context Protocol and optimized to give AI agents the right context without the bloat. Learn more in my beginner friendly deep dive on context engineering and the goldfish problem: https://lnkd.in/eDTYGWR7
-
Confused by Tree-sitter’s “named,” “fields,” and “children”? We used data from 25 languages to translate them into clearer ideas: Category, Thing, and two kinds of Connection, with roles and simple counts. The post shares the findings, a translation guide, and the code—so teams ramp faster and AI agents get better context This is the first of our Clarity Engineering series for developers. https://buff.ly/GqZmsQP
-
Your AI assistant is a goldfish 🐟 —it only sees what fits in the moment. We unpack how the context window really works, why stuffing it with too much (or the wrong) info makes answers worse, and what that means for getting reliable results. If you’ve watched a chat forget details or make things up, this one’s worth a read. Part 2 of our Intro to AI and the Economics of AI https://lnkd.in/eazet2xQ
-
-
Part 1 of our series on the fundamentals of generative AI, and the hidden economics that control it. https://lnkd.in/edje_WDZ
You don’t pay for words when you use AI. You pay for tokens — and for the GPUs crunching those tokens in real time.
-
We're excited to launch our first online presence -- knitli.com is live! 🎆 Check it out and let us know what you think.
-
Our founder, Adam Poulemanos, wrote about his decision to leave a successful career to start Knitli. Check out our origin story: https://lnkd.in/ej6jvi5y (he's also our only employee who likes to talk like there are others)