---
title: "Visual Explainers — page 5"
description: "Visual introductions to the engineering concepts behind the tools we use."
---

# Visual Explainers — page 5
> Visual introductions to the engineering concepts behind the tools we use.

## Visual Explainers

### [Apache Lucene Scoring: BM25 Relevance vs StableTfl Consistency](https://yomu.fyi/visual-explainers/apache-lucene-scoring-bm25-relevance-vs-stabletfl-consistency.md)
- Type: Comparison
- Tags: [Apache Lucene](https://yomu.fyi/topic/apache-lucene.md), [Benchmark](https://yomu.fyi/topic/benchmark.md), [BM25](https://yomu.fyi/topic/bm25.md), [Latency](https://yomu.fyi/topic/latency.md), [RAG](https://yomu.fyi/topic/rag.md), [Search](https://yomu.fyi/topic/search.md)

A comparison of Apache Lucene's BM25 scoring and StableTfl for paginated search: BM25 maximizes relevance, while StableTfl trades a small relevance drop for consistent ranking across replicas and pages.

### [The bfcache Trade-off: Instant Restores vs Silent Misses](https://yomu.fyi/visual-explainers/the-bfcache-trade-off-instant-restores-vs-silent-misses.md)
- Type: Trade-offs
- Tags: [bfcache](https://yomu.fyi/topic/bfcache.md), [Caching](https://yomu.fyi/topic/caching.md), [Latency](https://yomu.fyi/topic/latency.md), [Performance](https://yomu.fyi/topic/performance.md), [Progressive Web Apps](https://yomu.fyi/topic/pwa.md), [Service Workers](https://yomu.fyi/topic/service-workers.md)

bfcache freezes page snapshots for instant back/forward navigation, but unload listeners, cache headers, or third-party scripts can silently disable it. This explainer maps the trade-offs and the metrics that reveal them.

### [Argo CD: The Reconciliation Loop Between Git and Your Cluster](https://yomu.fyi/visual-explainers/argo-cd-the-reconciliation-loop-between-git-and-your-cluster.md)
- Type: What is it?
- Tags: [Architecture](https://yomu.fyi/topic/architecture.md), [Argo CD](https://yomu.fyi/topic/argocd.md), [Deployment](https://yomu.fyi/topic/deployment.md), [GitOps](https://yomu.fyi/topic/gitops.md), [Kubernetes](https://yomu.fyi/topic/kubernetes.md), [OpenShift](https://yomu.fyi/topic/openshift.md)

Argo CD is a GitOps controller that treats Git as the desired state and continuously reconciles it against what actually runs in each cluster. It maps the parts of that loop and why pull-based agents change its scale.

### [The API Gateway Boundary: Where Protocol, Policy, and Protection Meet](https://yomu.fyi/visual-explainers/the-api-gateway-boundary-where-protocol-policy-and-protection-meet.md)
- Type: Architecture
- Tags: [API](https://yomu.fyi/topic/api.md), [API Gateway](https://yomu.fyi/topic/api-gateway.md), [Microservices](https://yomu.fyi/topic/microservices.md), [Observability](https://yomu.fyi/topic/observability.md), [Rate Limiting](https://yomu.fyi/topic/rate-limiting.md), [Routing](https://yomu.fyi/topic/routing.md)

The API gateway is the single boundary where protocol translation, auth, quotas, and routing become shared policy instead of per-service code. This layered diagram shows what belongs there and why.

### [Streaming ASR vs. Batch ASR: The Buffer-Latency Trade-off](https://yomu.fyi/visual-explainers/streaming-asr-vs-batch-asr-the-buffer-latency-trade-off.md)
- Type: Comparison
- Tags: [ASR](https://yomu.fyi/topic/asr.md), [Latency](https://yomu.fyi/topic/latency.md), [Privacy](https://yomu.fyi/topic/privacy.md), [Real-time](https://yomu.fyi/topic/real-time.md), [Speaker Diarization](https://yomu.fyi/topic/speaker-diarization.md), [Streaming](https://yomu.fyi/topic/streaming.md)

Streaming and batch ASR pipelines compared: how input buffer length, speaker diarization, and where audio is processed trade latency against transcript quality and privacy.

### [Hybrid Agent Runtime: Antigravity SDK's Local and Cloud Split](https://yomu.fyi/visual-explainers/hybrid-agent-runtime-antigravity-sdk-s-local-and-cloud-split.md)
- Type: Architecture
- Tags: [AI Agents](https://yomu.fyi/topic/ai-agents.md), [AI Gateway](https://yomu.fyi/topic/ai-gateway.md), [Antigravity SDK](https://yomu.fyi/topic/antigravity.md), [Local AI](https://yomu.fyi/topic/local-ai.md), [Model Routing](https://yomu.fyi/topic/model-routing.md), [Privacy](https://yomu.fyi/topic/privacy.md)

Antigravity SDK's hybrid architecture separates planning from execution: a cloud model plans, local models run most tokens on-device. See how routing, execution, and the data boundary fit together.

### [How a 3D CT Model Reads a Volume: From Voxel Grid to Structured Report](https://yomu.fyi/visual-explainers/how-a-3d-ct-model-reads-a-volume-from-voxel-grid-to-structured-report.md)
- Type: How it works
- Tags: [3D CT](https://yomu.fyi/topic/3d-ct.md), [DICOM](https://yomu.fyi/topic/dicom.md), [Healthcare](https://yomu.fyi/topic/healthcare.md), [Machine Learning](https://yomu.fyi/topic/machine-learning.md), [Multimodal](https://yomu.fyi/topic/multimodal.md)

A walkthrough of how a 3D CT model turns a voxel volume into a structured diagnostic report: isotropic resampling, 3D patch tokens, spatial encoding, and language-model reasoning.

### [Blender's MCP Connector: Batch Speed vs. Authority Over the Scene](https://yomu.fyi/visual-explainers/blender-s-mcp-connector-batch-speed-vs-authority-over-the-scene.md)
- Type: Trade-offs
- Tags: [AI Agents](https://yomu.fyi/topic/ai-agents.md), [Blender](https://yomu.fyi/topic/blender.md), [Human-in-the-Loop](https://yomu.fyi/topic/human-in-the-loop.md), [Least Privilege](https://yomu.fyi/topic/least-privilege.md), [MCP](https://yomu.fyi/topic/mcp.md), [Tool Calling](https://yomu.fyi/topic/tool-calling.md)

Blender's MCP connector lets an LLM analyze and batch-edit scenes through Blender's Python API. This trade-offs view weighs that speed against the authority and isolation it requires.

### [How BioNeMo Turns a Target List Into Folded Structures](https://yomu.fyi/visual-explainers/how-bionemo-turns-a-target-list-into-folded-structures.md)
- Type: How it works
- Tags: [Benchmark](https://yomu.fyi/topic/benchmark.md), [BioNeMo](https://yomu.fyi/topic/bionemo.md), [GPU](https://yomu.fyi/topic/gpu.md), [Inference](https://yomu.fyi/topic/inference.md), [Life Sciences](https://yomu.fyi/topic/life-sciences.md), [NVIDIA](https://yomu.fyi/topic/nvidia.md)

BioNeMo Inference Runtime runs one pipeline per target: parse, tokenize, generate features, infer on a GPU replica, and write a structure. This diagram traces those stages and the throughput trade-off.

### [AI Data Assimilation: Turning Scattered Signals Into One State](https://yomu.fyi/visual-explainers/ai-data-assimilation-turning-scattered-signals-into-one-state.md)
- Type: Architecture
- Tags: [AI Data Assimilation](https://yomu.fyi/topic/ai-data-assimilation.md), [Architecture](https://yomu.fyi/topic/architecture.md), [CorrDiff](https://yomu.fyi/topic/corrdiff.md), [NVIDIA Earth-2](https://yomu.fyi/topic/nvidia-earth-2.md), [Ontology](https://yomu.fyi/topic/ontology.md), [StormCast](https://yomu.fyi/topic/stormcast.md)

AI data assimilation fuses a model's prior state with fresh, uneven observations into one consistent state. See the four-layer architecture and where each layer can break.

### [Binary Authorization: Block Unsigned Images or Just Log Them?](https://yomu.fyi/visual-explainers/binary-authorization-block-unsigned-images-or-just-log-them.md)
- Type: Comparison
- Tags: [Authorization](https://yomu.fyi/topic/authorization.md), [Binary Authorization](https://yomu.fyi/topic/binary-authorization.md), [Grafeas](https://yomu.fyi/topic/grafeas.md), [Kritis](https://yomu.fyi/topic/kritis.md), [Kubernetes](https://yomu.fyi/topic/kubernetes.md), [Supply Chain](https://yomu.fyi/topic/supply-chain.md)

Binary Authorization blocks container images that lack the attestations a cluster policy requires. This comparison sets a strict blocking posture beside an audit-only one, and shows how to choose between them.

### [How to Solve an Assignment Problem: Five Checkpoints from Spec to Solver](https://yomu.fyi/visual-explainers/how-to-solve-an-assignment-problem-five-checkpoints-from-spec-to-solver.md)
- Type: How to
- Tags: [Assignment Problems](https://yomu.fyi/topic/assignment-problems.md), [Decision-Making](https://yomu.fyi/topic/decision-making.md), [Load Balancing](https://yomu.fyi/topic/load-balancing.md), [Optimization](https://yomu.fyi/topic/optimization.md), [Scheduling](https://yomu.fyi/topic/scheduling.md)

A five-checkpoint workflow for assignment problems: write the spec, name the objective, test whether choices interact, pick a solver class by scale, then validate against live state.

### [Antibodies as a Prediction Pipeline: Where AI Models Break Down](https://yomu.fyi/visual-explainers/antibodies-as-a-prediction-pipeline-where-ai-models-break-down.md)
- Type: Trade-offs
- Tags: [Antibodies](https://yomu.fyi/topic/antibodies.md), [Biology](https://yomu.fyi/topic/biology.md), [Decision-Making](https://yomu.fyi/topic/decision-making.md), [Evaluation](https://yomu.fyi/topic/evaluation.md), [Life Sciences](https://yomu.fyi/topic/life-sciences.md), [LLMs](https://yomu.fyi/topic/llm.md)

Antibody discovery is a three-stage prediction pipeline, and AI models that excel on familiar targets degrade sharply on unfamiliar ones. This explainer maps where that gap appears and what it costs.

### [BIMI: The Identity Proof Behind the Logo](https://yomu.fyi/visual-explainers/bimi-the-identity-proof-behind-the-logo.md)
- Type: What is it?
- Tags: [BIMI](https://yomu.fyi/topic/bimi.md), [Deliverability](https://yomu.fyi/topic/deliverability.md), [Digital Identity](https://yomu.fyi/topic/digital-identity.md), [DMARC](https://yomu.fyi/topic/dmarc.md), [DNS](https://yomu.fyi/topic/dns.md), [Email](https://yomu.fyi/topic/email.md)

BIMI is a DNS-published claim that ties a verified brand logo to an authenticated domain, so mailbox providers can display it. Here is what each part of that claim requires.

### [BigQuery in the Right Lane: Five Checkpoints for AI Data Pipelines](https://yomu.fyi/visual-explainers/bigquery-in-the-right-lane-five-checkpoints-for-ai-data-pipelines.md)
- Type: How to
- Tags: [Apache Iceberg](https://yomu.fyi/topic/apache-iceberg.md), [BigQuery](https://yomu.fyi/topic/bigquery.md), [Cost Optimization](https://yomu.fyi/topic/cost-optimization.md), [Data Engineering](https://yomu.fyi/topic/data-engineering.md), [Data Pipelines](https://yomu.fyi/topic/data-pipelines.md), [Data Warehousing](https://yomu.fyi/topic/data-warehousing.md)

A five-checkpoint workflow for keeping BigQuery in the right lane: land and shape data once, tune partitions and history windows, and federate instead of copying.

### [Pass@1 vs. pass@k: Two Ways to Read One Benchmark](https://yomu.fyi/visual-explainers/pass-1-vs-pass-k-two-ways-to-read-one-benchmark.md)
- Type: Comparison
- Tags: [Benchmark](https://yomu.fyi/topic/benchmark.md), [Cost Optimization](https://yomu.fyi/topic/cost-optimization.md), [Latency](https://yomu.fyi/topic/latency.md), [Model Evaluation](https://yomu.fyi/topic/model-evaluation.md), [Model Routing](https://yomu.fyi/topic/model-routing.md), [Performance](https://yomu.fyi/topic/performance.md)

A benchmark can support two different verdicts: pass@1 on the first attempt, or pass@k with retries measured as cost per solved task. This comparison shows which reading matches your workload.

### [Agent Memory: What Outlives the Run](https://yomu.fyi/visual-explainers/agent-memory-what-outlives-the-run.md)
- Type: What is it?
- Tags: [Agent Memory](https://yomu.fyi/topic/agent-memory.md), [AI Agents](https://yomu.fyi/topic/ai-agents.md), [Databases](https://yomu.fyi/topic/databases.md), [LLMs](https://yomu.fyi/topic/llm.md), [Postgres](https://yomu.fyi/topic/postgres.md), [Retrieval-Augmented Generation](https://yomu.fyi/topic/retrieval-augmented-generation.md)

Agent memory is the knowledge that survives a run, not the working state inside one. This explainer separates the two, then maps the write path, the read path, and the database limits that cap recall.

### [Choose a Beach Like an Engineer: A Constraint-First Field Guide](https://yomu.fyi/visual-explainers/choose-a-beach-like-an-engineer-a-constraint-first-field-guide.md)
- Type: How to
- Tags: [Beaches](https://yomu.fyi/topic/beaches.md), [Decision-Making](https://yomu.fyi/topic/decision-making.md), [Latency](https://yomu.fyi/topic/latency.md), [Optimization](https://yomu.fyi/topic/optimization.md), [Routing](https://yomu.fyi/topic/routing.md)

Use a constraint-first workflow to choose among beach destinations: start with access, then shoreline character, activities, and crowding so the final pick matches the trip you actually need.

### [Bazel in the Inner Loop: From Monorepo Graph to Focused Sync](https://yomu.fyi/visual-explainers/bazel-in-the-inner-loop-from-monorepo-graph-to-focused-sync.md)
- Type: How it works
- Tags: [Architecture](https://yomu.fyi/topic/architecture.md), [Bazel](https://yomu.fyi/topic/bazel.md), [Caching](https://yomu.fyi/topic/caching.md), [Developer Experience](https://yomu.fyi/topic/developer-experience.md)

See how Bazel’s dependency graph can separate build scope from IDE scope, helping engineers move through large monorepos without syncing every module.

### [Authorization Is a Runtime Decision](https://yomu.fyi/visual-explainers/authorization-is-a-runtime-decision.md)
- Type: How it works
- Tags: [AI Agents](https://yomu.fyi/topic/ai-agents.md), [Auditability](https://yomu.fyi/topic/auditability.md), [Authentication](https://yomu.fyi/topic/authentication.md), [Authorization](https://yomu.fyi/topic/authorization.md), [OAuth2](https://yomu.fyi/topic/oauth2.md)

See how an agent request becomes an allowed or denied tool action through identity context, policy checks, scoped credentials, and an auditable decision trail.

[Newer explainers](https://yomu.fyi/visual-explainers/page/4.md) · [Older explainers](https://yomu.fyi/visual-explainers/page/6.md)