kadi
⚖️ Kadı — a calibrated judge in front of your frontier LLM. Routine decisions answered locally with a conformal-prediction guarantee (≤ α disagreement with the oracle); everything else escalated and learned from. Zero-dependency Python.
Get picks like this daily. The day's top launches, AI/tech news, and a weekly opportunity spotlight — straight to your inbox.
This is 1 of 65 launches in llm evaluation and deployment tools — see how it stacks up on momentum and crowding →
17 other launches read as similar to this one →
Details
- External ID
- 1413451466
- Source
- GITHUB
- Company
- —
- Product
- kadi
- Website domain
- github.com
- Launched
- Oct. 10, 2026
- Cohort
- —
- Upvotes
- 18
- Upvotes percentile
- 0.39
- Tags
- adaptive-conformal-inference, ai-agents, anthropic, claude, conformal-prediction, distribution-shift, gdpr, guardrails, knowledge-distillation, kvkk, llm, llm-cost-optimization, llm-routing, mlops, model-cascade, online-learning, privacy, python, uncertainty-quantification, zero-dependency
Enrichment
- Niche
- llm evaluation and deployment tools
- Vertical
- Horizontal
- Function
- Model & infra
- Audience
- Developer
- AI stance
- AI-native
- Project type
- Hobby / open-source project
- Normalized one-liner
- calibrated decision judge for frontier llms
- Manually corrected
- False
Could you build this?
Partial While the zero-dependency Python wrapper is concise, implementing valid statistical conformal prediction algorithms with mathematical finite-sample error guarantees requires specialized statistics knowledge.
What it would actually take: The implementation relies on conformal prediction theory to calibrate confidence scores and determine non-conformity measures for small local models vs. frontier oracles. The hard part is the theoretical statistical calibration ensuring bounded error rate (alpha) across arbitrary non-exchangeable real-world LLM query distributions. It requires expertise in statistical machine learning and uncertainty quantification.
Competitors
Other products that read as similar to this one — 17 launches clear the similarity bar, closest 8 shown.
Attention rank: #14 of 18 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 187 days after the earliest competitor.
- OpenDecision · github · 2026-09-17 · 47 upvotes · similarity 0.34
- judged.systems · ph · 2026-10-08 · 71 upvotes · similarity 0.33
- reflex · github · 2026-09-17 · 101 upvotes · similarity 0.33
- jevals · hn · 2026-09-20 · 47 upvotes · similarity 0.32
- DECIDE · ph · 2026-09-08 · 1 upvotes · similarity 0.32
- SQLazy · ph · 2026-09-21 · 2 upvotes · similarity 0.32
- ST-jeved · github · 2026-09-20 · 30 upvotes · similarity 0.32
- decider · github · 2026-09-16 · 130 upvotes · similarity 0.31
Other launches for this product
- No other launches for this product.