Live answer, requirement 5 of 12

Request classification and routing

I build automation that classifies each inbound request, routes it by role, and holds risky changes for human approval.

The requirementBuild automation that reads inbound work, decides what happens next, and acts under human oversight.

Try it

Paste an inbound request, such as a support ticket, and a demo model returns the same JSON contract my pipeline uses. The page shows the parsed fields, the routing decision, and whether the task would wait for human approval.

How it works

Map the inbound flowI sample real requests with the team and agree on the labels, the required fields, and the route for each label. Every sample enters through the public intake API on api.onluca.ai, POST /api/requests, which returns a stable request id.
Pin the output schemaI write the strict JSON contract first: intent, kind, risk, effort estimate, suggested role, and approval flag. A regex fast path labels the obvious requests, and I call the model only when confidence falls below 0.75.
Validate the outputI check every field before routing: risk must match its allowed values, and effort estimates are clamped to a fixed range. A failed model call falls back to medium risk with human approval required, so it fails closed.
Plan and routeThe dispatch loop classifies the request, then the planner I build writes a bounded task envelope. Each envelope names the agent role, the paths it may touch, and the checks it must pass. Read-only questions close as an answer instead of opening work.

Tools

Tool	What it does here	Status
Agent dispatcher and planner	Captures each inbound request, classifies intent and risk into the JSON contract, then writes bounded task envelopes for workers.	Early access
Agent runtime and tool bridge	Runs one routed task inside its envelope, calls only the tools that role allows, and reports status back.	Available
CRM sales automation engine	Runs event-triggered workflows with conditional steps and templates, and stores each step result, error, and duration.	Available
Policy engine approvals and audit	Parks a gated action until a named approver with the required capability decides, then writes that decision to the audit trail.	Early access
Demo classifier on Workers AI	Returns the same JSON fields my pipeline uses, so a visitor sees real classification output without an account.	Demo only

Available means shipped and in use. Early access means built but not generally available. Roadmap means designed, not built. Demo only means it powers this public demo, not customer deployments.

The full write-up covers access control, evaluation and how this runs inside a customer environment.

Read it on gusit.de

The other answers

LLM applications in productionEvery model call goes through one gateway, guardrails run on both sides, and each task gets a measured model choice.Prompt engineeringI write the output shape first, version the prompt, budget its tokens, and compare model replies before shipping.Retrieval-augmented generationGrounded, cited answers over the customer's own documents, with access rules applied before retrieval.AI agents and orchestrationI build agents that call your systems as registered tools, and the workflows that trigger and sequence them.AI workflow orchestrationI design the steps, contracts, and data flows, then ship them as running workflows on customer systems.API and integration designI publish the API contract first, gate it with keys and scopes, then build integrations against it.Hybrid enterprise integrationI connect AI services to your identity, applications and data, then run them where your policy allows.Prototype to productionI take a prototype through generated test suites and merge review gates to a zero-downtime deploy that rolls back on failure.Approvals and exception handlingI put a person between the agent and the change, with gated tools, a pending queue, expiry, and a record.Evals and output validationI score answers against a versioned rubric, trace failing calls to the integration that broke, and tune latency per call.Client-facing deliveryI turn engineering work into updates that client architects, security leads and business owners can each act on.

Gus IT LLC. Contact gus@gusit.de. Claude is a product of Anthropic PBC. Cloudflare and Workers AI are trademarks of Cloudflare, Inc. Llama is a trademark of Meta Platforms, Inc. Other names are trademarks of their respective owners. Gus IT LLC is an independent engineering firm and is not affiliated with or endorsed by these companies.