Labs
How TGL actually builds and operates.
TGL develops through a fleet of specialized, AI-assisted operating environments. Each owns a domain — development, communications, production, warehouse, Apple platforms — and each maintains its own operational history of the work it does.
The fleet
Six environments, five running.
ARCHIE
LiveDevelopment & New Applications
Generalist development environment. Application development, testing, deployment pipelines, prototypes, and new product work.
ADAM
LiveMail & Communications
TGL's communications infrastructure: mail systems, sending authentication and signing, and the services that depend on them.
EVE
LiveApplications & Production
Production applications, customer platforms, identity systems, hosting, and gateway infrastructure.
FILLY
LiveWarehouse & Fulfillment
Physical warehouse operations — receiving, restocking, shipping, label printing, and the hardware on the floor.
MAC
LiveApple Platforms
Development and deployment of TGL's iOS and macOS software.
SETH
In DevelopmentAutonomous Intelligence
Intended to become the intelligence and orchestration layer across the ecosystem — learning generalized operational patterns and coordinating the environments above.
Operational memory
Every environment keeps a record of its own work.
Where appropriate, TGL's environments preserve detailed local operational histories: development processes, commands run, code changes, errors encountered, the corrections that followed, deployments, tool activity, and results.
That record is operationally useful on its own — it is how a fix found once does not have to be rediscovered. It also represents potential governed research input for future SETH development.
To be clear: SETH has not been trained on this information. These histories exist; using them as training input is a future possibility subject to governance, not something that has happened.
What gets recorded
- Development process and decisions
- Commands and tool activity
- Code changes and deployments
- Errors, diagnoses, and corrections
- Operational outcomes
Why it matters
A mistake made once and understood is worth more than a process document nobody reads. The fleet's histories are how operational knowledge survives across time and across environments — and how a correction on one system becomes a rule everywhere.
Infrastructure
What TGL runs.
- Production
- Business platforms serving live operators across distribution, wellness, hospitality, wealth management, and events — hosted and operated on TGL infrastructure.
- Identity
- A central OAuth2 / OpenID Connect provider issuing authentication across TGL platforms.
- Communications
- TGL-operated mail infrastructure with per-domain mailboxes, sending authentication, and deliverability monitoring — plus managed device provisioning.
- Data & retention
- Operational databases per environment, scheduled encrypted backups, and off-box deep storage retention.
- Edge
- A reverse proxy layer terminating TLS for every TGL-operated domain, with default-deny network policy behind it.
- Monitoring
- Continuous health, uptime, and usage monitoring, with incident history surfaced to customers rather than kept internal.
R&D
What we're working on.
Labeled honestly. Research is research until it ships.
SETH
In DevelopmentIn-house intelligence and orchestration layer. Conversational business modeling, capability selection, and generation with human review.
In-house AI compute
In DevelopmentTGL-operated compute so intelligence workloads run on infrastructure TGL controls rather than rented capacity.
Capability generation
RoadmapAutomated generation of new library components from requirements, gated on human review before entering the permanent library.
Interested in how this is built?
We talk to technical people who find infrastructure more interesting than frameworks.