H
Howardism
Howardism · Vol. 03Plate I · No. 01

Writing, in order.

Pieces447Sections4Oldest10 Apr 2026Newest15 Sept 2026

Every article in the wiki, grouped by kind: Concept notes, Entity profiles, and Essay pieces. Hover any title for a preview; click to enter.

Filed under
allConceptEntityEssayIndex
Domain
agent systemsagent securityai coding practiceevals & benchmarksmodel capability & trainingalignment & safetyinterpretabilityinteraction & multimodalformal mathstartup & founderproduct & orgai economics & laborsuperintelligence trajectoryentities
Plate · 01 of 04
299
concept notes

Concept, in order.

Evals & Benchmarks22
  1. 01
    Adaptive Stopping in Evaluation Sampling
    Capability EvaluationBenchmarksEvaluation Methodology+3
    · 22′
  2. 02
    Evaluation-Time Answer Leakage
    BenchmarksEvaluation MethodologyConstruct Validity+3
    · 40′
  3. 03
    Skill Lift
    EvaluationAgent SkillsAblation+3
    · 20′
  4. 04
    AI-Assisted Error Analysis
    EvalsEvaluationError Analysis+3
    · 16′
  5. 05
    GDPval Benchmark
    EvaluationBenchmarksEconomic Impact+1
    · 34′
All 22 →
Plate · 02 of 04
81
entity notes

Entity, in order.

  1. E01
    DHH (David Heinemeier Hansson)
    EntityPersonOpen Source
    · 7′
  2. E02
    Robert C. Martin (Uncle Bob)
    EntityPersonSoftware Craft
    · 8′
  3. E03
    Bridgewater Associates
    EntityOrganizationFinance+1
    · 8′
  4. E04
    Aakanksha Chowdhery
    EntityPersonResearcher+2
    · 21′
  5. E05
    Azalia Mirhoseini
    EntityPersonResearcher+2
    · 15′
  6. E06
    CS329A: Self-Improving AI Agents (Stanford)
    EntityDocumentCourse+2
    · 54′
  7. E07
    US Center for AI Standards and Innovation (CAISI)
    EntityOrgAI Evaluation+2
    · 5′
  8. E08
    Greptile
    EntityCode ReviewVendor
    · 4′
  9. E09
    Transluce
    EntityOrgAI Evaluation+2
    · 6′
  10. E10
    LlamaIndex
    RetrievalDocument ParsingAgent Engineering+1
    · 5′
All 81 entity notes →
Plate · 03 of 04
51
essay pieces

Essay, in order.

  1. S01
    Does the Augmentation/Automation Split Govern Skill at Work?
    DerivedWorkforceHuman Capital+2
    · 22′
  2. S02
    Memory-Poisoning Numbers, Conditioned on the Write
    DerivedSecurityMemory+1
    · 26′
  3. S03
    Is Persistence the Line Between Prompting and Spec-Driven Development?
    DerivedAI Coding WorkflowPlanning+1
    · 18′
  4. S04
    Writer/Reviewer vs Agent-to-Agent Review
    DerivedCode ReviewAI Coding Workflow+2
    · 27′
  5. S05
    The Configuration Gap: What an Assurance Argument Measures vs What It Covers
    · 24′
  6. S06
    The Price of Mixing Agents, and the Principal Nobody Counted
    · 26′
  7. S07
    Safety Commitments That Cannot Bind the Actor Who States Them
    DerivedGovernanceAI Policy+2
    · 20′
  8. S08
    The Data Wall and the Validation Commons Are One Supply Constraint
    DerivedData WallVerification+3
    · 22′
All 51 essay pieces →
Plate · 04 of 04
16
index notes

Index, in order.

  1. I01
    Evals & Benchmarks
    · 12′
  2. I02
    Open Questions Dashboard
    · 3′
  3. I03
    Open Questions Backlog
    · 102′
  4. I04
    Interpretability
    · 5′
  5. I05
    Agent Security
    · 20′
  6. I06
    Agent Systems & Harness Engineering
    · 29′
  7. I07
    AI Coding Practice
    · 23′
  8. I08
    AI Economics & Labor
    · 14′
All 16 index notes →
Plate · #
425
subjects

By subject, most cited.

Agent EngineeringEntityAnthropicAlignmentEvaluationGovernanceLLM ArchitectureAI Coding WorkflowEmpiricalSecurityDerivedSafetyPersonHarnessHuman AI CollaborationWorkforceBenchmarksMeasurementTest Time ComputeInterpretabilityStartupCapability EvaluationContext ManagementEconomicsMonitoringPost TrainingAI Native OrgLLM ModelOpenaiProduct ManagementCode ReviewEvaluation MethodologyFounderMulti AgentMultimodalLLM As A JudgeOpen WeightsOrgPlanningPrompt InjectionRecursive Self ImprovementAgent OrchestrationCode QualityEngineering MetricsFailure ModesThreatsAI PolicyCapability TrajectoryClaudeCostKnowledge ManagementReinforcement LearningAI CodingGoogleOpen SourcePromptingReasoningReward HackingTheoryVerificationAI NativeAI RdForecastingGovernance WorkforceOrg DesignReference MonitorResearcherZero TrustAccountabilityAgentsAI For MathematicsClaude CodeCoordinationCybersecurityEvalsFrontier LabsIncident ResponseInference ScalingMCPMemoryOrganizationReliabilityRetrievalReward DesignSynthetic DataTechnical DebtTool UseTrainingTrust BoundaryAdaptive EvaluationAI EvaluationAI LabArchitectureAuthorizationAutomationChain Of ThoughtHonestyIdentityLLM EvaluationMethodologyMoatsModel WelfareOptimizationOrchestrationPractitioner OpinionProduct OrgProduct StrategyRole EvolutionSearchSupply ChainTasteTeam DesignToolTraining GamingWorkflow DesignAgent DeploymentAgent HarnessAgent RuntimeAgentic RlAI AdoptionAsiAuthenticationCalibrationCode GenerationCodingCoding AgentsCognitive LoadCompetitive StrategyConstruct ValidityDeceptionDeep ResearchDesignDeveloper WorkflowFormal MethodsGo To MarketIntegrationLeast PrivilegeMacroMisalignmentOversightPersonaPrdProductProduct ProcessPropagationPrototypingRspScalingSelf ReportSkillsSoftware EconomicsStanfordSuperintelligenceTechnology DiffusionTestingThreat LandscapeType/entityAccess ControlAI CoworkingAI Native OrganizationBelief ModificationCareerCatastrophic RiskCodexCognitionCompanyCompetitive LandscapeComputeCoworkData ContaminationData WallDeepmindDefense In DepthDefensibilityDeploymentDocumentDocument ParsingDual UseEducationEngineering LeadershipExpertiseGame TheoryGrowth DynamicsHallucinationHRHuman CapitalInferenceInterfaceInterpretability ResearcherIntrospectionLaborLimitsLLM CharacterMental ModelMetrModel ImprovementModel RoutingModel SpecPlatformPrivacyProcessProduct ArchitectureProduct TasteProductionProductivityProtocolProvenanceQualityQuantizationResearch MethodologyResearch ProgramRLHFSoftware ArchitectureSoftware DesignStandardsStartupsSynthesisSystemTask SpecificationUniversal AIVendorVendor ClaimVerifiabilityVulnerability ResearchY Combinator
+200 one-off subjects
AblationAbstractionAgencyAgent ArchitectureAgent FrameworksAgent SafetyAgent SecurityAgent SkillsAgentic EvaluationAgileAIAI EconomyAI EducationAI EthicsAI SafetyAI ToolsAixiAlignment ResearcherArms ControlArtifactsAsset PricingAuditingAuthorshipAutonomyBenchmarkBenchmark IntegrityBusiness StrategyCapabilityCausal TheoryCharacterClassifiersCLI AgentCode AuthorshipCoding AgentCollaborationCompetitionComplexity TheoryComputer UseConcept DiscoveryConfirmation BiasConfused DeputyConsciousnessContainmentContradictionControl PlaneCooperationCopyrightCourseCovert ChannelCreativityCredentialsCredit AssignmentCultureData FlywheelData PipelineData ProvenanceDebate MapDebuggingDecision MakingDefenseDefinition Of DoneDefinitionsDelegationDelimiter InjectionDemocratizationDesign PrincipleDesign ProcessDesign SystemsDeterminismDeveloper EndpointDiffusionDigital IntelligenceDiscoveryEconomic ImpactEconomic ModelingEconomistEfficiencyEmergent BehaviorEngineering IntelligenceEnterprise DeploymentEpistemicsEquityError AnalysisExploit DevelopmentFailure ModeFaithfulnessFinanceFormal VerificationFoundersFrontier LabGeneralistsGeneralizationGoodhartGoogle DeepmindGrounded TheoryGroundingGroup AgencyGrpoGtmHciHiringHistoryHouseholdHuman In The LoopIncentivesIncidentsInference EfficiencyInformation Flow ControlInfrastructureInput ValidationIntelligence MeasureInteraction MultimodalInternal ToolingInverse ScalingIsolationJailbreakKnowledge SharingLatent CapabilityLearning GuideLeverageLifecycleLLM CapabilitiesLoopsLow RankMarket DesignMatrix CompletionMeasurement ValidityMechanistic InterpretabilityMediaMemorizationMessaging PlatformMethodMetricsMicrosoftMidtrainingModelModel CapabilityModel PoisoningMultilingualMvpNational SecurityNetflixNous ResearchNvidiaOauthObservabilityOff HostOpen Ended DiscoveryOpinionsOwaspPermissionsPersonal Knowledge BasePlatform EngineeringPm SkillPricingPrincipleProbesPrompt EngineeringPsychometricsQualitative AnalysisRAGRed TeamingRepresentationsResearchResearch AgendaRevenue Per EmployeeReward ModelingReward ModelsRl Post TrainingRole ShiftRubricsSafeguardsSandboxingScale StageScaling LawsScientific DiscoverySFTSingularitySkills DevelopmentSoarSoftwareSoftware CraftSoftware DevelopmentSoftware ParadigmStatisticsSubagentsSuperpositionSystemsTalent DensityTeam DynamicsThreat ModelTool PoisoningTraining EfficiencyUncertaintyUnit EconomicsUnlearningValidationValue ModelVersioningWorkflow