Docs
OfficialJev Practices & Patterns
Patterns
Architectural patterns for System One systems: speculative fan-out, confidence-gated routing, composite scoring, and intent routing.
- Format
- Docs
- Reading time
- 2 min
- Published by
- TypeSafe AI
- Last verified
- Sep 2026
Highlights
- The index lists four patterns: Speculative Fan-Out, Confidence-Gated Routing, Composite Scoring, and Intent Routing.
- Fan-out's triage example sends five questions in one request and routes on thresholds like bug severity > 1.5 or refund probability > 0.7.
- Confidence-gated routing's voice-banking example sends anything below 0.6 to a support agent and needs above 0.85 to approve a transfer.
- Composite scoring normalizes four 5-level Scores by dividing by 4, then applies role weights like 40% Python and 40% system design for a senior IC.
- Intent routing pairs an intent Choice with a complexity Score, escalating when complexity is above 1 or its confidence falls below 0.5.
Quickstart
python
category = response.answers["category"]
bug_severity = response.answers["bug_severity"]
refund = response.answers["refund_requested"]
if category.choice == "bug_report":
if bug_severity.score > 1.5 and response.answers["has_reproducible_steps"].noul > 0.6:
escalate_to_engineering(ticket_id, severity="high")
elif category.choice == "billing":
if refund.noul > 0.7:
route_to_billing_with_flag(ticket_id, refund_likely=True)Watch out
The index only summarizes the four patterns, and each pattern page's thresholds are examples, so retune them to your own stakes and traffic.
Questions about this entry
Short answers drawn from the linked project or page.
Four: Speculative Fan-Out, Confidence-Gated Routing, Composite Scoring, and Intent Routing, each with its own page. The index says every pattern page's thresholds are examples, so retune them to your own stakes and traffic.
Fan-out asks everything in one request and filters in code; confidence-gated routing sets per-action thresholds; composite scoring normalizes several Scores and applies weights; intent routing pairs an intent Choice with a complexity Score.
Fan-out is filed under Cost and Speed; confidence-gated routing under Reliability and Safety; composite scoring under Cost, Reliability, and Speed; and intent routing under Cost and Speed.
No. It is an architectural overview page on the docs site; each pattern links to its own page, and no package or repository is offered here.
They assume you already know the TypeSafe primitives and understand how confidence works, and the index says to read those pages first if you do not.
The page invites readers who have found a use case they think should be listed to drop the team a note, saying they are keen to learn how people use the primitives.
Thinking in discrete, atomic decisions that compose into complex system behavior, which the page calls a key skill for getting the most out of TypeSafe.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: How the agent stack moved in Jev's first three days
YouTube: Diogo Almeida tells swyx what System One is, why RLHF failed his automation goal, and why AI should eventually disappear.
X: The 10-step roadmap, summarised
X: Giving your agents a decision brain
More Jev resources like this
Ask everything the system might need in one request, then let code throw away what it does not use. The ticket-triage example sends a category plus four speculative questions in parallel.
Jev Practices & PatternsDocs#official#patterns#batching
Official
Classify intent with a Choice and complexity with a Score, then send each branch to deterministic code, a specialist model, or a person, with a 0.5 intent-confidence floor.
Jev Practices & PatternsDocs#official#patterns#routing
Official
Instructions, option descriptions, score levels, and noul criteria all accept JSON, so you can label question parts, pass schemas and taxonomies, and keep candidate paths alive.
Jev Practices & PatternsDocs#official#primitives#architecture
Official
Use the answer to decide what to do and confidence to decide whether to act: the voice-banking example routes below 0.6 to a human and needs 0.85 or more to auto-approve a transfer.
Jev Practices & PatternsDocs#official#patterns#confidence
Official
The idea behind Jev: models that return typed decisions and probabilities instead of generated text, named after Kahneman's fast-thinking System 1, with confidence to route around uncertainty.
Jev Guides & ArticlesDocs#official#docs#architecture
Official
A maintained list of jev-1.13's known failure modes, literal reading, unreliable counting, dates as text, indirection, context rot, and contradictory criteria, each with a guardrail.
Jev Practices & PatternsDocs#official#docs#evaluation
Official
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.