BoundaryML drops feelings for typed AI if-statements

// 1h agoOPENSOURCE RELEASE

BoundaryML drops feelings for typed AI if-statements

BoundaryML has launched feelings, an open-source project demonstrating how semantic evaluations and typed AI if-statements can be implemented natively in BAML rather than requiring a bespoke language. By pairing BAML with TypeSafe AI's Jev classifier, the library introduces methods like .feels() and .matches<T>() for fast probabilistic routing while dispatching generative tasks to LLMs via .ask().

// ANALYSIS

Probabilistic "AI if statements" do not need bespoke toy DSLs when robust type systems and dedicated classification models can simply turn vibes into first-class methods.

• Pragmatic language design: Rather than inventing a whole new programming language for fuzzy evaluation, feelings implements .feels() in roughly 20 lines of BAML through a blanket interface implementation over all types.

• Specialized compute routing: Offloading fast semantic classifications and routing decisions to TypeSafe AI's Jev while reserving heavyweight generative LLMs for .ask() dramatically lowers latency and token costs.

• First-class developer tooling: Developers gain static analysis, compile-time exhaustiveness checking on pattern matching, offline testing, and shell-friendly parallel streaming (such as semantic grep) out of the box.

• Strict runtime guardrails: The system explicitly restricts classifier return shapes to booleans, floats, enums, and structured classes, preventing malformed calls before any network requests occur.

// TAGS

feelingsbamlboundarymltypesafe-aijevai-developmentllmdevtoolopen-sourcetype-safetyprompt-engineering

DISCOVERED

1h ago

2026-09-19

PUBLISHED

1h ago

2026-09-19

RELEVANCE

8/ 10

AUTHOR

vaibcode

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

UPDATE22m ago

Synara teases redesigned model picker

Synara, an open-source, local-first desktop workspace designed for orchestrating AI coding agents, previewed an upcoming redesign of its model picker interface. Built to unify runtimes such as Claude Code, Codex, Cursor, and Gemini CLI into a single control plane, Synara uses the model picker to let developers switch between local providers, models, and reasoning effort levels while preserving session context during agent handoffs.

OPEN SOURCE1h ago

Chunklaya unlocks long context for Laya

chunklaya is an open-source Apache 2.0-licensed harness designed to scale Laya, an open-weights System 1 decision model, into an effective local alternative to TypeSafe's proprietary Jev model for long documents. To overcome Laya's decision-head opening bias past 200 tokens, chunklaya splits text into discrete passage chunks, replaces uncalibrated default queries with descriptive prompts to cut false positives by 25x, and uses length-sorted batching to run efficiently on consumer hardware.

OPEN SOURCE2h ago

Sotto debuts as an open-source, local-first macOS dictation tool powered by Whisper and Apple Silicon MLX models.

Sotto is an open-source, local-first dictation application for macOS that transcribes speech directly into active text fields upon releasing a push-to-talk hotkey. Built around a decoupled architecture, the tool pairs a native Swift menu bar client with a Bun-compiled TypeScript and Fastify inference server that orchestrates Whisper for speech recognition and MLX-accelerated language models for automatic punctuation and text cleanup. The application runs completely offline on Apple Silicon Macs while also supporting remote inference on separate Mac or Linux machines. Showcased by developer Ben Davis to explore how AI coding agents enable multi-language systems development, Sotto delivers a privacy-respecting, subscription-free alternative to proprietary voice-to-text utilities.

BoundaryML drops feelings for typed AI if-statements