Platform runtime
Clustering. Peer cluster with gossip membership.
Sharding. Deterministic entity partitioning.
Entity rebalancing. Moves on membership change.
Durable execution. Workflows resume from last step.
State replication. Entity state across zones.
Multi-region. Active-active state across regions.
Split-brain resolution. Pick a partition strategy.
HA / DR. No single point of failure.
Rolling updates. Zero-downtime version rollout.
Auto-scaling. Cluster size follows load.
Self-healing. Failed members replaced automatically.
Backpressure. End-to-end flow control.
Work-stealing. Idle members pick up queued work.
Gossip protocol. Peer-to-peer membership.
Failure detection. Phi accrual, tunable sensitivity.
Cross-region failover. Reroutes on region degrade.
Session stickiness. Requests routed to entity shard.
In-memory grid. State in RAM for microsecond reads.
Akka Actors
Actor model. Unit of state, behavior, and one mailbox.
Serialized execution. One message at a time.
4KB actors. Memory footprint of an idle actor.
200M actors per core. Density on a single CPU core.
1M-node clusters. Cluster size the runtime is built for.
Actor hierarchy. Parents supervise their children.
Supervision. Resume, restart, stop, escalate.
Backoff supervision. Restarts spaced by backoff.
Death watch. Termination arrives as a message.
Typed behaviors. Protocols checked at compile time.
Tell and ask. Fire-and-forget, or request with reply.
Mailbox types. Bounded, priority, control-aware.
Stashing. Messages held until behavior changes.
Dispatchers. Thread-pool, pinned, affinity pools.
Actor timers. Single-shot and periodic messages.
Receive timeout. Fires when an actor idles too long.
Routers. Pool and group, round-robin or hashed.
Location transparency. Same reference on any node.
Remoting. Actor messaging over encrypted transport.
Cluster singleton. One instance across the cluster.
Distributed pub-sub. Topic delivery across the cluster.
Circuit breaker. Trips on failure, half-opens to retry.
Coordinated shutdown. Ordered phases on exit.
Stream integration. Actors as sources and sinks.
Development model
Agent component. LLM unit with session memory.
Workflow component. Durable multi-step orchestration.
Event-sourced entity. State from event log.
Value entity. State as latest value.
View component. Materialized projection over events.
Endpoint component. HTTP, gRPC, or MCP surface.
Consumer component. Subscribes to events or topics.
Timed action. Scheduled durable work.
Component composition. Typed message passing.
Java 21+. Java 21 language and above.
Component annotations. Declarative wiring.
Effect API. Value plus side effects.
Akka console. One interface, from build to cost.
Testing framework. Unit and integration kits.
Maven plugin. Scaffolding, build, packaging.
SBT plugin. Same for Scala projects.
Hot reload. Code changes without restart.
CLI. akka command for deploy, logs, exec.
IDE integration. IntelliJ and VSCode plugins.
Serialization. Protobuf, JSON, custom codecs.
Component versioning. Schema evolution rules.
Akka console
System. The system and its regulatory role.
Agents. Every agent, Akka or not, in one directory.
Agent identity and cost. Models, tools and cost.
Ask. Questions answered from live records.
Build. Specifications, code, exit conditions and tests.
Agent playground. Try any model with a judge.
Orchestration. Every component of a running service.
API endpoints. Call an endpoint and see its response.
Controls. Risk, regulations, signed controls.
Activity. Conversations, traces and trace replay.
Human. Decisions, changes under review and halts.
Evidence. Verdicts, compliance and evidence packs.
Cost. Spend, budgets, changes, models and training.
Data & state
Event sourcing. Immutable event per state change.
Event journal. Durable queryable event log.
Snapshotting. Periodic state snapshots.
CQRS. Separate command and query paths.
Materialized views. Event-driven read models.
In-memory state grid. State cached across cluster.
Per-entity consistency. Commands run in order.
Cross-entity consistency. Async projections.
At-least-once delivery. Retry with idempotency keys.
Exactly-once effects. Effect ledger dedupes retries.
CRDT support. Distributed Data replicated types.
Key-value entity. Non-event-sourced state.
Schema evolution. Versioned events and states.
Agentic AI
Agent runtime. Session-scoped LLM with retry.
Session memory. Durable per-session context.
Multi-agent orchestration. Teams as workflow steps.
Tool binding. MCP consumption from any component.
Tool exposure. Any component becomes an MCP tool.
Agent-to-agent (A2A). Cross-service agent RPC.
Streaming responses. Token-level to clients.
Structured outputs. Schema-validated JSON.
Function calling. Typed argument marshalling.
Model provider abstraction. Swap models per agent.
Prompt templates. Versioned reusable prompts.
Context management. Auto compaction, summary.
RAG integration. Retrieval before generation.
Embeddings. Vector generation and storage.
Semantic search. Vector, keyword, hybrid.
Trace propagation. End-to-end agent trace.
Agent identity. Each model call names its agent.
Model support
Anthropic Claude. Claude 5, 4.8, 4.7, Haiku 4.5.
OpenAI. GPT-5 and GPT-4o families, o-series.
Google Gemini. Via AI Studio API.
Google Vertex AI. Enterprise-hosted variant.
Fireworks. Hosted open-weight models.
AWS Bedrock. AWS multi-model service.
Azure OpenAI. Azure-hosted GPT.
Self-hosted. Gemma, Llama, Qwen, Mistral.
Local inference. Ollama and vLLM endpoints.
Custom endpoints. Bring-your-own HTTP or gRPC.
LoRA adapter serving. Task adapters over a base.
Model routing. Per-request policy selection.
Inference routing
AI gateway router. Routing decisions a proxy applies.
Agent-blind routing. Agents never name a model.
Two-stage pipeline. Rules first, then meaning.
Rulesets. Named rules enforced as one set.
Predicates. Equals, prefix, suffix, contains, in, exists.
No regex predicate. Closed set, no request-path stall.
Protocol-agnostic. One rule across providers.
Prompt-text matching. Whole prompt, or last user turn.
Header matching. Any request header, lower-cased.
Closed destination set. Undeclared targets refused.
Immutable revisions. Revision numbers never reused.
Conditional writes. If-Match and ETag on every write.
Ruleset history. Every change, revision, and author.
Rule dry run. Try a request before shipping the rule.
Classifier registry. Models a rule may consult.
Immutable classifiers. A changed prompt is a new id.
Classifier deferral. A rule asks a model to decide.
Conversation pinning. Later turns follow the first model.
Embedding match. Vector arithmetic, no model call.
Zero-token decisions. Semantic routing, no tokens.
Discovery mode. Traffic observed, nothing routed.
Routing mode. A manifest routes unmatched requests.
Embedder identity gate. Blocks a changed embedder.
Frozen route header. Hashed into the manifest.
Tag bound once. A tag never points elsewhere later.
Evidence log. Input, time, and rule version per decision.
Exemplar privacy gate. Rarely seen text withheld.
Retention opt-out. Drop the stage, retain no text.
Capture stages. Summary line, local file, object storage.
Gap markers. Unrecorded exchanges, named.
Content never logged. Only a recorder exports bodies.
Model serving
vLLM serving. Open-weight models on your own GPUs.
Model descriptor. One file deploys a model.
Descriptor validation. Checked before deploy.
Descriptor export. Live deployments back to file.
Accelerator classes. GPUs carved into capacity.
Dedicated or shared. Cards for one model or many.
VRAM budget. The share of a card per model.
GPU inventory. Kind, memory, free count, per node.
Cards per node. A multi-card model needs one machine.
Portable GPU names. One descriptor, any region.
Tensor parallelism. One model split across cards.
Pipeline parallelism. One model split across stages.
Data parallelism. Replicated weights across cards.
Expert parallelism. Experts sharded across cards.
Prefill/decode split. A fleet per phase.
Continuous batching. Batch limits per deployment.
Chunked prefill. Long prompts split into steps.
Quantization. Weight precision set per deployment.
KV-cache dtype. Cache precision set per deployment.
CPU offload. Host memory holds weights past VRAM.
Swap space. Host memory backs the KV cache.
LoRA adapter serving. Adapters over a shared base.
Adapter by request. A base:adapter name picks one.
Adapter discovery. Found by storage prefix.
Queue-depth autoscale. Replicas follow the queue.
KV-cache autoscale. Replicas follow cache pressure.
Model rollouts. Surge and availability bounds.
Inference routes. Models mapped to hostnames.
Model proxy. A local port onto a deployment.
Tool-calling parser. Per-model parser and template.
Max model length. Context ceiling set per deployment.
Hugging Face source. Token and weight cache.
Akka Optimize
Observe. Measure work agents do on live traffic.
Route. Send each task to the fitting model.
Train. Specialize models on your own data.
Serve. Run in your environment, frontier fallback.
Targeting engine. Scores narrowness, volume, spend.
Deterministic grader. SQL and JSON outputs.
LLM judge. Scores unstructured prompts.
External harnesses. Copilot, Claude Code, Cursor.
Governance signal. Scores drive audit trail.
Training signal. Scores drive reward model.
Promotion signal. Scores drive promotion.
Open-weight base tuning. 4–12B base with RL.
LoRA adapter. ~50MB weights you own.
In-VPC training. GPUs in your cloud account.
Air-gapped training. Supported where required.
Fireworks training. Alternative to in-VPC.
Quality gate. Quality ≥ baseline before promotion.
Token gate. Token drop ≥ target before promotion.
Shadow staging. Candidate runs alongside incumbent.
Canary promotion. Gradual rollout by traffic share.
Fleet promotion. Full rollout after gates clear.
Before/after benchmark. Saved to audit record.
Post-promotion grading. Catches drift live.
Auto-retrain on drift. Fresh candidate to staging.
Auto-revert on slip. Return to proven model.
Improvement policy. Bounds tuning scope and targets.
Interaction log. Every decision written durably.
AI Gateway. OpenAI-compatible ingress.
Tier select. Per-request model tier.
Adapter hot-swap. Live swap without restart.
Response cache. Gateway-level caching.
Gateway rollback. Canary, shadow, rollback controls.
Frontier fallback. GPT-5.5, Opus 4.8 as teacher.
Trace store. Agent traces retained.
Scrub & curate. PII and secret redaction.
Judges & evals. Reward-model pipeline.
RL environments. Sandbox training suites.
Synthetic data. Teacher and RLVR generation.
SFT · DPO · RL. Three training regimes.
Elastic inference. Base plus adapters, auto-scale.
BYO-agent plugin. Endpoint config, trace capture.
Log forwarder. Prod logs to gateway.
Repo & CI connection. Codebase, sandbox deploys.
90-day assessment. Install, Observe, Witness.
Reinforcement learning. Trained with TRL and verl.
Engine by GPU count. Engine fits the GPUs given.
Smoke check. A real slice trained before full spend.
Smoke verdicts. Six outcomes, ungraded included.
Pause and resume. A run stopped and started again.
Durable checkpoints. Candidates outlive their job.
Per-run cost budget. A ceiling set before a run starts.
Training-job logs. Read without cluster access.
Content-hashed datasets. Same bytes, one dataset.
Separate evaluation set. Scoring data kept apart.
akka optimize. Optimize commands in the Akka CLI.
akka models. Model serving from the Akka CLI.
Spend. Where the money goes, and why it moved.
Token efficiency. Waste ranked, safest first.
A saving, explained. Evidence, estimate, change.
Use cases. Spend by use case, found in captured traffic.
Budgets. Burn and forecast per team or app.
Variance explained. Volume, mix, tokens and price.
Change preview. Cost and quality before shipping.
Staged rollouts. A change rolled out behind a stop rule.
Realized savings. Actual against estimate.
Cost–quality frontier. Cheapest model that qualifies.
Model prices. Listed, negotiated or allocated.
Model registry. Hosting and governance per model.
Model lifecycle. Promote, deprecate, retire, sign off.
Artifact registry. Versioned datasets and models.
Self-hosted break-even. GPUs against a hosted API.
Prompt-cache tokens. Cache reads counted in cost.
Akka Verify
Batch evaluations. Scheduled offline runs.
Online evaluations. Continuous live scoring.
Guardrails. Input and output policy enforcement.
Policy records. Machine-readable per decision.
Evidence export. Regulator-ready audit bundles.
Per-decision audit trail. Inputs, outputs, policy.
SAFR mapping. Safeguards for agentic finance.
Regulation corpus. 227 regulations and frameworks.
Explainability records. Reasoning trace per output.
Drift detection. Behavior-change alerts.
PII detection. Auto-redaction hooks.
Prompt-injection defence. Layered input filters.
Model comparison. A/B judging harness.
Red-team harness. Adversarial test suite.
Tool permission. Tool names against an allow list.
Tool correctness. Tools used against expected.
Argument correctness. Against expected arguments.
Latency budget. Recorded latency against a duration.
Token budget. Tokens used against a limit.
Model-call budget. Model calls against a limit.
Retrieval metrics. Recall, precision, MRR, NDCG@K.
Faithfulness. Reply against retrieved passages.
Plan metrics. Quality, adherence, efficiency.
Bring-your-own judge. Your model, our prompt.
Rubric override. Your rubric beats the built-in id.
Judge-call cap. A ceiling on judge calls per run.
Checkpoint evaluation. Trained models scored first.
Tool-call evaluation. An agent's requests, scored.
Risk survey. The risks the system carries.
Controls draft. Accept, waive or reject controls.
Signed bundles. Sign controls, compare versions.
Controls in force. Each control and whether it held.
Control drift. Which controls breach or changed.
Verdict drift. Verdict changes by bundle.
Control evidence. The evidence behind a verdict.
Compliance. Compliance now and over time.
Evidence packs. Signed evidence to hand an auditor.
HITL console. Decisions waiting on a person.
Changes under review. Stage, evidence, approvals.
Halts. Who stopped a capability, and why.
HITL quality. How well people oversee the AI.
Conversation transcripts. Each turn with its verdicts.
Governance ledger. SPIFFE and trace ids per record.
Guardrail SPI. Custom guardrails, tool calls too.
Evaluator components. Each bound to a control id.
Classifiers. Content classified before a policy applies.
Scoped sanitizers. Redaction within a declared scope.
Oversight records. Every review and halt, recorded.
Retention. Records kept for the period a regulation sets.
Legal hold. Records frozen against deletion.
Adversarial testing
Adversarial campaigns. Attacks on your own system.
AILuminate v1.1 hazards. Results by hazard.
OWASP LLM Top 10. A second classification.
Three-verdict scoring. Broke, held or inconclusive.
Cost-ordered tiers. Rules, heuristics, then a judge.
Secret exfiltration. A withheld marker quoted.
Prompt injection. The injected marker read back.
PII leak detection. Cards, SSNs, your identifiers.
Jailbreak scoring. No refusal, task fulfilled.
Guardrail scoring. Reply checked against policy.
Refusal consistency. An early refusal reversed later on.
Jailbreak judge. A model reads undecided cases.
Attack-goal judge. Decides if an attack succeeded.
Attack wrappers. Base64, ROT13, language, role play.
Adaptive attackers. PAIR, Crescendo, Tree of Attacks.
Versioned corpora. Dated sets with attribution.
Bring-your-own judge. Your model, our prompt.
Breaks and holds. Two counts, never a single rate.
Akka Specify
Dev Spec. Goals, requirements, knowledge sources.
Eval Matrix. Risk, operational, improvement envelopes.
Goals & intent. Purpose, success metrics, autonomy.
Functional requirements. What the system must do.
Knowledge sources. What the system must know.
Risk envelope. Safeguards it must not cross.
Operational envelope. SLA, regions, latency, cost.
Improvement policy. What it should optimize toward.
Versioned specs. Every spec tracked with history.
Build track. PM, Dev, ML/Data Eng, Analyst.
Govern track. Risk, Security, Compliance, SRE, FinOps.
Continuous build. Spec changes regenerate code.
Continuous governance. Eval Matrix in every build.
Spec-driven delivery. Features specified first.
Akka plugin. 22 commands in your AI harness.
Install from akka.ai. One line, 20 tested harnesses.
Exit conditions. When a feature is done.
Definition of done. Every exit condition and its state.
Conform verdict. READY_TO_SHIP or NOT_READY.
Ship targets. Review or release, with ship steps.
Conformance modes. Enforced or À la carte.
Enterprise setup. CI, scanning, supply chain.
Converge. Unbuilt work added to the task list.
Resilience tests. Against live endpoints.
Inspect. Endpoints, state and UI of a running service.
Generated docs. Checked for language and tone.
Self-documenting. Diagrams generated from the code.
Self-explaining code. Docs for every endpoint.
Integration tests. Every testkit test and its last result.
Evals. Agents evaluated before they ship.
Adversarial tests. Agents attacked before they ship.
Integration & protocols
HTTP endpoints. REST, streaming, SSE.
gRPC endpoints. Bidirectional protobuf streams.
MCP tool binding. Consume MCP from components.
MCP tool exposure. Expose components as MCP.
Agent-to-agent (A2A). Cross-service agent RPC.
Brokerless messaging. Component pubsub in-process.
Kafka bridge. Consume and produce topics.
Pulsar bridge. Pulsar equivalent.
CDC ingestion. Debezium-style change streams.
Server-sent events. SSE for streaming clients.
WebSocket. Long-lived client connections.
Deployment & operations
Regions. Multi-region deployment.
Geo-pinning. Traffic and residency by region.
Rolling updates. Zero-downtime rollout.
Blue/green. Parallel-version staging.
CI/CD hooks. Actions, GitLab, Jenkins.
Secrets management. Encrypted, runtime-injected.
Network policies. Service-to-service rules.
CVE patching. Managed cadence, opt-in windows.
Deploy from CLI. akka deploy per project.
Deploy from console. Click-through from AAO.
Deploy from API. Programmatic control.
Project isolation. Per-project quota, fences.
Environment isolation. Dev, stage, prod.
Resource limits. CPU, memory, concurrency caps.
Identity & access
RBAC. Four built-in roles.
OIDC. Federated login via any provider.
SAML. Enterprise SSO integration.
Project membership. User-to-project mapping.
Environment isolation. Grants per environment.
API tokens. Long-lived and short-lived.
Service accounts. Non-human identities for CI.
MFA. Multi-factor for console access.
Role-change audit log. Every grant recorded.
Session management. Lifecycle and revocation.
Security & compliance
In-VPC deploy. Runtime in your VPC.
In-account deploy. Runs in your cloud account.
Air-gapped. No internet dependency.
Sovereign. Customer controls all data plane.
TLS 1.3. Encryption in transit everywhere.
AES-256 at rest. Encryption for persisted state.
Key rotation. Managed or bring-your-own KMS.
Audit logging. Control-plane actions logged.
SOC 2 Type II. Annual audit, NDA report.
ISO 27001. Annual certification.
GDPR. DPA and subprocessor list published.
HIPAA readiness. BAA available on request.
DPIA support. Assessment inputs supplied.
Subprocessor list. Public, notified changes.
Observability
Distributed tracing. Per-request across services.
Agent trace. Per-invocation LLM trace.
Workflow trace. Step-by-step execution trace.
Metrics export. Prometheus and OTLP.
Event-journal query. Per-entity time-series.
Component health. Live status per component.
Log aggregation. Structured JSON with context.
Grafana integration. Dashboards and alerts.
Datadog integration. APM, metrics, logs.
Prometheus integration. Native scrape endpoint.
OpenTelemetry. Traces, metrics, logs on OTel.
Structured logs. Schema-aware format.
