Every generation rebuilt the stack. Mainframes gave way to client-server. Client-server gave way to cloud. Now AI is rewriting everything again, not as a feature bolted onto Gen3, but as a fundamentally new architecture where agents are first-class citizens and applications drive themselves.
Monolithic, batch-processed, operator-driven. Centralized compute, dumb terminals, rigid processes.
PCs, GUIs, distributed databases. Software you install. The ERP era: SAP, Oracle, PeopleSoft.
Multi-tenant, browser-based, API-driven. Configuration over code. The platform era.
Self-driving applications, self-learning systems. AI agents as first-class operators. The entire stack rewritten for autonomy.
Active Projects
ZebraDB
Hybrid transactional-analytical data platform for agent-native applications.
ZFlow
Durable execution, event streaming, and code runtime.
ZSL
Terraform for enterprise applications. Declarative, plannable, deployable.
CENTARI
A relational foundation model for the enterprise.
ENT-Bench
Application-layer benchmarks for evaluating AI agents on real enterprise tasks.
TetrisUI
Composable, modular application runtime. Software that upgrades and maintains itself.
Vapor
DOSS app builder platform. Composable, generatable applications from ZSL declarations.
Tapestry
DIY federated context graph. A composable knowledge fabric that weaves enterprise context across systems, tenants, and time.
MetaHarness
Self-learning, self-evolving agent runtime with pluggable model-serving. The infrastructure layer under dossbot.
DossGrid
High-performance data grid SDK with zero-copy Arrow Flight SQL. 100 rows to 100 million.
Platform architecture
Gen4 requires interlocking systems built together from first principles. Foundation models, a system of record, a schema language, orchestration, action, and composable applications, each reinforcing the others. DOSS builds them all together.
Legacy vendors have the system of record but no foundation models. AI-native companies have intelligence but no operational database. Bolting one onto the other creates friction. DOSS builds both together.
| Foundation models | System of record | System of schema | Orchestration | System of action | Applications | |
|---|---|---|---|---|---|---|
| Oracle | — | ● | ○ | ○ | ○ | ● |
| SAP | — | ● | — | ○ | ○ | ● |
| Salesforce | ○ | ● | ○ | ● | ● | ● |
| Palantir | ● | — | ● | ● | ● | ○ |
| ServiceNow | ○ | — | ○ | ● | ● | ● |
| DOSS | ● | ● | ● | ● | ● | ● |
Data -> models
Enterprise operations generate structured training data. CENTARI learns from real schemas, workflows, and process logs, not synthetic benchmarks.
Models -> systems
Foundation models improve every layer. Better schema inference, smarter workflow optimization, predictive resource allocation, autonomous operations.
Systems -> data
Improved systems handle more complex operations, generating richer training data. The flywheel compounds: each cycle makes the next one more valuable.
Hard problems
15 scored problems across infrastructure and intelligence. The substrate that makes Gen4 possible, and the frontier research that makes it revolutionary.
Composable Data Platform
A hybrid transactional-analytical database that adapts its physical schema per tenant, supports real-time OLAP without ETL, and dual-writes every mutation as cell-level AI training events.
Agentic HTAP
ZebraDB.ai. A database that agents can clone, build on, test against, and deploy to production through the same typed API that humans use.
Hi-TX Durable Execution
A workflow engine where every execution is durable, retriable, and observable, and every step emits an OCEL 2.0 event that feeds the foundation models.
Meta-Ledgers
Everything is a ledger. Every state change, every transaction, every decision, recorded in append-only ledgers that serve as both audit trail and training corpus.
ZSL: System of Schema
Terraform for enterprise applications. A single declaration fans out to data, logic, and applications simultaneously. The configbase IS the codebase.
CI/CD of the Configbase
The Agentic Application SDLC. The configbase has a filesystem (ZSL files), a compiler (static analysis), a test suite (generated UTs), a deployment pipeline (topological migration), and version control (branching/merging).
Context Christmas Tree
Semantic cache invalidation. The context tree grounds foundation model predictions in the actual state of the business, not stale snapshots.
Cell-Level Tokenization
The CENTARI Data Log. Every mutation to a Z-Table is simultaneously written as an append-only cell-change event with causal context, bridging the system of record and the foundation models.
Foundation Models for Enterprise Data
The Large Tabular Model. A relational foundation model trained on cell-change events with graph-aware attention, producing predictions about the state of the enterprise.
Enterprise World Modeling
The Large Enterprise System Model. A dynamics model that answers 'what happens next?' and 'what happens if?' by simulating the future state of the enterprise.
Neuro-Symbolic Fusion
DDM as physics. Encoding business logic as machine-readable symbolic constraints that regularize neural training and guarantee constraint satisfaction at inference.
Cross-Tenant Transfer
The schema diversity moat. Foundation models that improve with every new customer because each tenant's ZSL schema contributes relational patterns the model has never seen.
Human-Agent Parity
The Centaur Runtime. AI agents operate through the exact same interfaces, constraints, permissions, and audit trails as human users. Same path, same rules, same accountability.
Self-Building Applications
Given a ZSL declaration, the platform generates the complete application: data model, UI, dashboards, APIs, workflows, validation, agent harnesses. A new module is a file, not a 6-month project.
Network Intelligence
EDI-NET. A data network that creates cross-entity, cross-company intelligence. Every trading partner adds nodes and edges to the CENTARI graph, enabling supply chain simulation across company boundaries.
