YouTube video by Cloudy with a Chance of Insights, The MSFT Podcast

Large language models mean not having to use complicated regular expression handlers to turn text into data. Using TypeChat, you can ensure that that data is type-safe JSON.

TypeChat is a library that makes it easy to build natural language interfaces using types. - GitHub - microsoft/TypeChat: TypeChat is a library that makes it easy to build natural language interfac...

TypeChat is a library that makes it easy to build natural language interfaces using types. - GitHub - microsoft/TypeChat: TypeChat is a library that makes it easy to build natural language interfac...

Das neue Release der Microsoft-Library für strukturierte Ausgaben großer Sprachmodelle ist weiterhin experimentell, bringt jedoch neue Funktionen.

Podcast Episode · Cloudy with a Chance of Insights · 10/20/2025 · 53m

For now, I tried a few lightweight checks in Colab; * * * I think the **constraint-first direction itself is quite reasonable**. In particular, I like the separation between a small human-readable intent representation and the deterministic/static/runtime checks around the LLM. There are related patterns in things such as Microsoft’s DSL Copilot and TypeChat: the formal representation helps, but practical reliability still comes from combining it with validation, diagnostics, and repair rather than expecting the model alone to behave like a compiler. After reading through IPL Studio and doing a couple of small controlled runs, my main thought is that it may help to separate **which freedoms IPL itself is supposed to eliminate** from the freedoms intentionally left to the architecture/code-generation stages. Something roughly like this: Natural-language requirement ↓ IPL / intent contract ↓ (optional normalized / typed authoritative IR) ↓ architecture / topology contract ↓ code lowering ↓ static gates ↓ runtime / behavioral verification ↓ repair That separation would let several different meanings of “deterministic” coexist without forcing IPL to solve all of them: Goal | Probably the important contract ---|--- Same requirement → same observable behavior | IPL + behavioral verifier Same requirement → roughly same architecture | explicit topology/architecture contract Preserve exact names/types/formulas | semantic-lowering contract Same input → identical source tree | deterministic lowering after an authoritative IR Repeatability across inference runs | model/backend/version/seed/runtime controls too So I would not necessarily make IPL larger or stricter first. My default route would be to **keep the minimalist vocabulary, but make the boundaries measurable**. The highest-information, relatively low-cost additions seem to be: 1. report **first-pass** , deterministic-repair, and LLM-repair results separately; 2. ensure the initial generator and behavioral oracle have the same requirement/fixture information; 3. measure **semantic preservation** separately from final runtime PASS; 4. measure architecture/topology variance separately from business behavior; 5. keep an equal-information direct natural-language baseline, mainly as a control rather than as a competitor to IPL. The third item in particular seems useful for this project. A final PASS/FAIL can currently combine several very different things: identifier / output-key preservation type preservation formula / arithmetic preservation control-flow preservation topology / entrypoint conformance cross-file dependency closure runtime behavior If those become separate receipts, then when generation fails you can say _which layer still had freedom_ , rather than only saying that the whole generated application failed. A very small Colab sanity check (click for more details) One benchmark distinction that may be worth making explicit (click for more details) Why I think an architecture contract could complement IPL rather than compete with it (click for more details) I would also consider an IPL → code semantic-preservation receipt (click for more details) Where this seems to sit relative to some existing work (click for more details) About the '13 verbs' question (click for more details) A small terminology/evaluation note on 'deterministic' (click for more details) If I had to pick only **one** next step, I would not start by adding more verbs or replacing the current pipeline. I would make the existing scorecard more layer-aware: 1. IPL parse / semantic-contract checks 2. topology / entrypoint checks 3. identifier + type + formula preservation 4. project integration / dependency checks 5. first-pass runtime behavior 6. deterministic repair 7. LLM repair after diagnostics Then the interesting claim becomes much easier to test: > **Which classes of freedom does the minimalist IPL remove, and which classes remain in the LLM-controlled stages?** To me, that is actually a stronger and more useful question than requiring IPL to make every generated source tree identical. The minimalist intent layer can remain minimalist, while the surrounding contracts progressively remove exactly the kinds of freedom that turn out to matter in practice.

Top LinksThe Hanselminutes Podcast – Introducing TypeChat...

Scott talks to Anders Hejlsberg about a new open-source p...