Joppa Technologies · Applied AI studio

AI that acts,
not just answers.

We engineer the harness and memory layer around frontier models, turning raw intelligence into agents you can depend on. Software that doesn’t just answer; it acts, with a human on the decisions that count.

Zora is live in production5 tools replaced by 1A human on every send
Our mission

We believe technology should give people back their most finite resource: their attention.

Not by replacing the work that matters, but by carrying everything around it, so a person who runs a business one relationship at a time is free to do the part only they can do.

Our first product, Zora, does this for event professionals: it reads the inbox, drafts replies, holds the calendar, and sends invoices on its own, with a human on every decision that counts. Beneath it: a dependable harness and a memory that compounds.

What we build

Agent equals model plus harness.

A frontier model is the smallest part of a working agent. The reliability lives in the harness around it, the memory beneath it, and the evals that prove it. That's what we engineer.

01

Agent harnesses

The runtime around the model: orchestration, control loops, tool use, with a guardrail on every step. What turns a clever demo into an agent that holds up in production.

02

Memory & retrieval

A memory layer with retrieval (RAG) so the agent remembers across sessions and grounds every answer in your real data, not the model’s guesses.

03

Evals & fine-tuning

We measure reliability with evals, then sharpen behavior with fine-tuning. What gets measured is what you can trust.

04

Systems of action

The whole workflow, inbox to invoice, shipped as a production app that plans, acts, and verifies, with a human in control.

Inside the harness

What we engineer around the model.

"Harness" isn't a metaphor. It's the runtime around a frontier model. The model is a small, stateless core; the harness gives it grounding, an act-and-verify loop, and a human on the send. Hover any block to see what each part does.

01

Context engineering

The system prompt, retrieval, and memory curated into a finite context window.

02

Tools & MCP

Permissioned actions over the Model Context Protocol: the agent’s hands.

03

Orchestration

Plans the work, runs the loop, fans out to sub-agents.

04

Verification

Rules, visual, or LLM-as-judge checks against a source of truth.

05

Guardrails

Hard limits and prompt-injection defense: the ‘never’ rules.

06

Human-in-the-loop

A person approves every consequential, irreversible action.

07

Observability

Every model call, tool call, and retrieval traced as one run.

08

Sandbox & cost

Least-privilege execution plus token budgets and prompt caching.

Inputs / grounding
Requestgoal in
Instructions
Memory & retrieval
Agent loop · gather → act → verify
Orchestratorgather · plan
Modelthe model
Tools & MCPact
Verify ↺check · repeat
Release gate
Guardrails
Human approvalyou, on the send
Action ✓
Cross-cutting
Observability
Sandboxing
Cost control
The harness: a model running in a loop
Hover any block to see what it does. A request is grounded by context, then the orchestrator runs the model through a loop (gather, act, verify), repeating until it’s right, before a verified action passes a human-approval gate. Observability, sandboxing, and cost control watch every step.
hover any component · the agent loops until it verifies · observability, sandboxing & cost control watch every layer
Our thesis

Any model can ace a demo. Only a harness survives production, so that's the part we build.

Proof of work

Zora: built, shipped, running.

Zora is a production system of action for event professionals: one tool replacing five, running the client workflow end to end.

Product · live in production

The AI back-office for event pros.

An agent harness over a memory layer: it reads the inbox, drafts quoted replies in the planner's voice (checked against their real pricing), manages the calendar, sends invoices, and gives each client a branded portal. One system where there used to be five.

5 → 1
Tools replaced
Memory layer
Per-business context
Verified
Every draft fact-checked
Human-in-loop
You approve every send
Let's build

Have a workflow worth automating?