InGoPenAIbyMinh Le Duc·Sep 27Jev’s Boring Use Cases Are the Ones That Actually Work💡 TL;DR & Key Takeaways:
InGoPenAIbyMinh Le Duc·Sep 26What Developers Actually Built With Jev in the First Week💡 TL;DR & Key Takeaways: Jev, a decision‑only AI model released on September 15 2026, was quickly adopted by developers: pi‑warden, built…
InGoPenAIbyMinh Le Duc·Sep 25What Is RLCD, and Why It Makes Jev Fast Instead of Smart💡 TL;DR & Key Takeaways: TL;DR RLCD is a proprietary training method from TypeSafe AI that teaches Jev to produce calibrated probability…
InGoPenAIbyMinh Le Duc·Sep 24Jev, Explained: The AI That Refuses to Write a WordTL;DR & Key Takeaways: TL;DR: Jev is a “System One Model” that returns typed, calibrated probabilities for predefined question types in a…
Minh Le Duc·Sep 23SmolVLM2 vs Qwen2.5-VL: Real-Time RTSP Edge Video Summarization Under 8GB VRAMRunning live RTSP video summarization under an 8GB VRAM limit requires predictable token economics. Here is how SmolVLM2 beats Qwen2.5-VL…
InGoPenAIbyMinh Le Duc·Sep 22Repo-to-Skill Framework Boosts Agent PerformanceRepo-to-Skill turns repos into skill contracts: higher agent success, fewer tokens.
Minh Le Duc·Sep 17Dynamic Reasoning Budgets Cut Inference CostsDynamic reasoning budgets route simple tasks fast and reserve capped reasoning for complex ones, cutting token use and latency by half.
Minh Le Duc·Sep 16Deterministic Agent Runtime with Context PruningLearn how a deterministic harness prunes context, uses a ledger and transactional tool calls to keep LLM agents reliable over many turns.
Minh Le Duc·Sep 5Agent harness or agent framework? They’re not the sameA framework is a library you build an agent with; a harness is a runtime that owns the loop. The difference is control flow, not features
InAI Advances BriefsbyMinh Le Duc·Sep 5Fun Project: I Built a Compressor That Thinks in TokensField notes from a weekend spent teaching an entropy coder to speak LLM.