Give your agents durable, governed memory — vectors, graphs, documents, and events in one engine. Built on the open-source ProximaDB context database and the Victor agent framework; run managed in your cloud (AKS today, ECS / GKE next) with tenant isolation, usage metering, and MCP built in.
No credit card · 1 GB storage · 0.5 GB ingest · 0.1 GB embedding · 10 TB scan · 1 connector
Designed for engineering teams building with AI agents in these categories
Agents start every session from zero, and teams lose weeks per quarter to tribal knowledge loss and repeated investigations — because nothing remembers what was already solved.
The same Kafka lag spike, Postgres connection pool exhaustion, or Kubernetes OOMKill gets investigated fresh every time it recurs — because the RCA from six months ago is buried in a comment thread nobody can find.
Your best engineers carry institutional knowledge no one else has. When they leave, go on vacation, or join a different team — that knowledge evaporates.
New engineers escalate everything because they don't know what's been solved before. Senior engineers spend hours on issues they've already fixed three times.
AnvaiOps connects to your existing tools and surfaces the right fix at the right moment — no workflow changes required.
Authorize AnvaiOps to read from beta sources such as Jira, Slack, Teams, Confluence, PagerDuty, Datadog, GitHub, and ServiceNow. Access is read-only by default, with real-time webhook ingestion available where configured.
AnvaiOps ingests your historical tickets, threads, and runbooks. We extract RCAs, fixes, and owning teams using AI — then store them as a searchable knowledge index.
When a new incident opens, engineers search in natural language. AnvaiOps returns relevant historical fixes, root causes, workarounds, and owning-team context with supporting metadata.
Open-source engine (ProximaDB), open-source agent framework (Victor), and a managed control plane that adds tenant isolation, governance, and metering on top.
Your on-call agent remembers every incident. 20+ connectors (Jira, Slack, PagerDuty, Confluence…) stream your operational history into tenant-isolated hybrid search — vector + BM25 + metadata filters, one query.
Your coding agent remembers your whole codebase. Code-property graphs + symbol-level embeddings (victor-codegraph), served to any agent over governed MCP.
Jupyter-style notebooks that run inside the per-workspace pod and connect to ProximaDB through an embedded ProximaSession. Markdown + code cells, syntax-highlighted Monaco editor, cell interrupt.
pgwire-compatible SQL bench wired directly to ProximaDB's federated SQL surface. Arrow result streaming, vector-aware result tables, automatic write-statement advisories.
Every workspace gets an embedded Gitea repo: notebook history, SQL queries, scripts, and config land as commits. Branch, diff, restore — even on a free trial.
Statuses below mirror ProximaDB upstream — see the pricing matrix for which tier ships each capability and RELEASE_TAXONOMY.md for the contract.
Canonical vector CRUD + similarity search over REST v2 / gRPC v2. Production SLA, full semver compatibility.
Tantivy-backed BM25 fused with vector similarity via RRF or weighted scoring. REST v1 hybrid facade today; gRPC parity in progress.
VECTOR_SEARCH, GRAPH_QUERY, DOCUMENT_QUERY, LOGS, METRICS table-valued functions in one SQL plane.
Node / edge CRUD, traversal, neighbors, shortest-path APIs. Native graph storage, not a relational fake-out.
Time-stamped tables with ASOF joins, downsampling, and OHLC aggregations. Designed for high-cardinality observability and trading workloads.
Ballista-orchestrated DataFusion MPP over cataloged Arrow snapshots. Available on Enterprise; Private Preview on Pro / Business.
Each account is bound to one cloud at signup (Azure, AWS, GCP). Data, workspaces, kernels, files — all in your tenancy, never traversing AnvaiOps infra.
One account can run workspaces in multiple regions within its cloud — workspace-scoped region picker on creation, immutable thereafter.
Bring-your-own Entra tenant. Multi-tenant app reg, app roles, scoped API audience, MFA enforced at the edge.
Tiered allowances on storage / ingest / embedding / scan with per-tier soft caps. Prometheus metrics for every meter; the calculator below mirrors the runtime gates.
Managed AnvaiOps SaaS, private VPC, or fully air-gapped on-prem. Three ProximaDB transports — REST, gRPC, Arrow Flight — fit isolated environments.
The original AnvaiOps workload: connect every incident to its resolution. Slack & Teams bots, 21 ITSM / observability / messaging connectors, hybrid retrieval over tickets and threads. Still a first-class use case, no longer the only one.
Code-graph as a service (in private preview): index your repositories into a correlated substrate — one symbol = one graph node + one vector + one relational row. Query impact_analysis (blast radius), transitive callers, and semantic fusion_search over code.
AnvaiOps mirrors ProximaDB's four-state release ladder. We never claim a status above the engine's upstream value — see ADR-0014 for the contract.
Production SLA, semver compatibility, full support contract. Vector CRUD & similarity search is the canonical Supported surface today.
Functional and API-stable; no production SLA yet. Federated SQL, hybrid retrieval, ORION graph, observability runtime, security runtime, TST time-series, event sourcing.
Scaffolded; partial implementation. Distributed cluster execution, full SQL parity, external tables (Iceberg / Delta / Hudi), DataFusion OLAP, Trino / Spark adapters.
Per-account grant for Beta or Experimental capabilities not included in your tier by default. Operator-driven for design partners; request access via the sign-up form.
AnvaiOps is powered by ProximaDB — an open-source multi-model database that unifies the indexes and data planes our knowledge layer depends on. The engine is public and auditable. The intelligence layer is what we build on top.
One engine for the multiple index types modern knowledge retrieval requires — unified under a single query plane. No separate search infrastructure to operate or pay for.
Server: standalone service — cloud, on-prem, or intranet-isolated. Managed: process lifecycle handled for you. Embedded: single-binary in-process for air-gapped environments. Same engine, three topologies.
Security-conscious enterprise buyers can inspect the open-source engine. No black-box indexing path, no hidden retention layer. Fork it, audit it, trust it.
No migration, no workflow changes. AnvaiOps reads from your tools and makes them smarter.
No search infrastructure to size. Connect your support stack, choose async or sync ingest, and pay for indexed knowledge, retrieval, and outbound transfer.
Capability badges below — GA Beta Experimental 🔒 Preview — follow our release taxonomy. We never claim a status above the engine's upstream value.
| Meter | Unit | Rate | What drives it |
|---|---|---|---|
| RetrievalPrimary | per TB scanned | $0.02 | You're billed for the GB of corpus your queries physically read — not the total size of your knowledge base. Intelligent indexing skips blocks that don't match your query filters before retrieval runs. Example: 5,000 queries/mo, 2 GB corpus, 40% scan fraction = 4,000 GB scanned/mo = 4 TB scanned = $0.08/mo. Tighten further by splitting your corpus into multiple named indexes (route runbook queries to your runbook index only) or adding metadata filters (team, severity, date) — both physically reduce scanned bytes. Team includes 30 TB/mo scan budget; Pro includes 300 TB/mo scan budget; Business includes 1 PB (1,000 TB)/mo scan budget. |
| Storage | per GB-month stored | $0.25 | Persistent searchable incident knowledge — indexes, metadata, and replicas in the primary region. Cross-region DR replication is billed only when selected or included by tier. Billed against the actual storage footprint so a 200KB Confluence page is billed fairly against a 20KB Jira ticket. This is managed searchable knowledge, not raw object storage. |
| Ingest / IndexingKIU | per GB indexed | $0.75 BYO vectors or document payload |
Write, validate, index, and compact customer-supplied content or vectors. v1/v2 vector batch APIs accept embedded vectors; v3 documents can also carry vectors through ProximaDB.Included ingest scales by tier from 1.5 GB on Team to 50 GB on Business. Ongoing retention is covered by the storage meter. |
| Managed EmbeddingKEU | per GB embedded | $5.00 $8.25 sync · BGE -25% · large $26.25 / $52.50 |
Optional embedding performed by AnvaiOps when you send documents without vectors. Async uses batched OpenAI text-embedding-3-small by default; hosted BGE models are available at 25% below the comparable OpenAI lane. Sync is available on Pro and above for immediate searchability. Small managed embedding allowances are included by tier. Native document ingest is $5.75/GB async when KIU + OpenAI KEU are used together, or $4.50/GB with hosted BGE. OpenAI text-embedding-3-large is available as a premium managed embedding option. Teams that bring precomputed vectors skip KEU. |
| Extra Connector | per connector / mo | $39 | Beyond the included connector count. Each adds a scheduled polling worker (every 2–12 hours depending on source). Real-time Slack and Teams webhook ingestion is included in the platform fee — not a per-connector charge. All connectors included on Enterprise. BYOC available for regulated or on-prem sources. |
| Network Egress | per billable GB | cloud nominal rate | Outbound transfer for result exports, cross-cloud access, on-prem access, and selected DR replication. Provider free allowances and same-cloud paths are reflected in the calculator. Network transfer is tenant-specific even when compute is pooled. |
Introductory rates use pooled compute to keep entry pricing low. Tenant data remains isolated and is billed through simple GB-based meters: retrieval (GB scanned), ingest/indexing (GB indexed), managed embedding (GB embedded), and storage (GB-month). No pass-through cloud bills. Dedicated Enterprise deployment starts at +$1,500/mo when isolation, private networking, or customer-controlled cloud is required.
Our support team uses AnvaiOps the same way your engineers will. Every ticket we resolve for a customer gets indexed back into our own instance, so our institutional knowledge compounds with each interaction. That lets us include support in the platform fee while keeping response quality high.
Start with a 30-day trial — 1 GB indexed storage, 0.5 GB ingest, 0.1 GB managed embedding, 10 TB scan budget, and one connector. If your agents (or your engineers) keep re-solving the same problems, we'll work with you directly and move you to a paid plan after the trial.