Docs
OfficialJev Practices & Patterns
Pattern: Composite Scoring
Break a judgment into atomic Score questions, normalize each by its top level, and combine them with weights you control in code, as in the resume example's four scored dimensions.
- Format
- Docs
- Reading time
- 2 min
- Published by
- TypeSafe AI
- Last verified
- Sep 2026
Works forJev for Scoring
Used inJev for HR & Recruiting
Highlights
- Resume screening asks four Score questions in one parallel request, each with five levels.
- Normalize each score by dividing by len(criteria) - 1 before weighting.
- Senior IC weights: 40% Python, 10% leadership, 40% design, 10% generalist.
- Engineering manager weights: 15% Python, 40% leadership, 20% design, 25% generalist.
Quickstart
python
py = response.answers["python_depth"].score / 4
lead = response.answers["team_leadership"].score / 4
arch = response.answers["system_design"].score / 4
general = response.answers["generalist"].score / 4
# Senior IC
ic_score = (0.40 * py) + (0.10 * lead) + (0.40 * arch) + (0.10 * general)
# Engineering Manager
em_score = (0.15 * py) + (0.40 * lead) + (0.20 * arch) + (0.25 * general)Watch out
The weights are policy, not model output, so revisit them whenever the role or the cost of a bad hire changes.
Questions about this entry
Short answers drawn from the linked project or page.
In your code, not the model. The resume example normalizes each score by dividing by len(criteria) - 1, then applies role-specific weights — 40/10/40/10 for a senior IC and 15/40/20/25 for an engineering manager.
So the policy stays adjustable without touching the model. Four parallel Score questions with five levels each are combined in code, and the page says to revisit the weights whenever the role or the cost of a bad hire changes.
Each dimension is normalized to 0-1, for example by dividing the score by 4, then multiplied by its weight, which keeps the nuance of each score while setting relative importance.
For a senior IC, 0.40 to python depth and system design, 0.10 to team leadership and generalist; for an engineering manager, 0.40 to leadership, 0.25 generalist, 0.20 architecture, 0.15 python.
Visibility into how the final score is calculated; if the top candidates do not match expectations you can adjust the weights to find the right balance.
Break a complex judgment into independent dimensions, score each one separately, and combine them with weights you control in code.
When you want to rank a set of items based on several criteria at once and need an adjustable, inspectable way to set each dimension's relative importance.
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
Rate content against 2 to 10 ordered levels. Returns a probability-weighted score that can land between levels, per-level probabilities, and confidence, with the arithmetic left to code.
Jev Practices & PatternsDocs#official#primitives#score
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
Architectural patterns for System One systems: speculative fan-out, confidence-gated routing, composite scoring, and intent routing.
Jev Practices & PatternsDocs#official#patterns#architecture
Official
Published evaluations of four automation workflows, security incidents, agent trace observability, invoice processing, and customer service, comparing Jev and frontier LLMs as structured workflows versus single prompts.
Jev Practices & PatternsDocs#official#benchmarks#evaluation
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
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
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.