Custom AI agent development
Agents built around how your team already works.
We design, build and ship AI agents that do real work inside your tools: reading documents, updating systems, drafting replies and making routine decisions, with a person approving anything that matters. Tested on your real cases before launch.
- 4 to 8 weeks per agent
- 4–8 wk
- From scope to live
- 100%
- Actions logged
- 0
- Irreversible steps without approval
In short
Velum's custom AI agent development designs and builds production AI agents for a specific business workflow. Each engagement covers agent design, integrations with your systems, human approval steps, an evaluation suite built from your real cases, a supervised pilot and launch, typically in four to eight weeks.
A good fit for
- Teams with a clear, repetitive workflow to hand off
- Companies that need agents inside existing systems, not a new app
- Enterprises that need approvals, audit and on-premise options
What you get
6 deliverables, all yours.
Everything below is handed over at the end, documented and owned by your team.
Scope this with us- 01
Agent design
What the agent reads, decides and does, what it must never do, and where a person approves.
- 02
Integrations
Connections to your CRM, ERP, help desk, email, chat and documents through APIs, MCP servers or browser automation.
- 03
Approval steps
Review and approve in Slack, Teams or email, with the evidence attached to every request.
- 04
Evaluation suite
Hundreds of test cases drawn from your real work, run on every change so quality never quietly slips.
- 05
Supervised pilot
The agent runs alongside your team on live work, with every action reviewed until it earns trust.
- 06
Handover
Runbooks, dashboards and training so your team owns the agent after launch.
How it runs
From first call to done.
- Week 1
Design
Workflow mapped, agent behaviour and limits agreed, test cases collected.
- Weeks 2–4
Build
Agent, integrations and approval steps built and evaluated against your cases.
- Weeks 4–6
Pilot
Supervised run on live work with the team, tuning until acceptance criteria are met.
- Weeks 6–8
Launch
Released as the default path, with monitoring, documentation and training.
Timeline
4 to 8 weeks per agent
Pricing
Fixed price per build, quoted against an agreed scope and acceptance criteria.
Where this applies
Solutions this service delivers.
- GTM engineering
Signal-based outbound
Outbound that starts with a reason to reach out.
A weekly pipeline of decision makers who are already showing the need you solve, researched, verified and ready for an outcome-led email.
- 15 h
- Research saved per SDR, weekly
- < 1 day
- From signal to first email
- < 2%
- Bounce rate, verified only
- Customer support
Support resolution desk
Tickets resolved, not deflected.
Two thirds of routine tickets resolved end to end with answers grounded in your policies and live account data, the rest routed with full context, and refunds or credits approved by a person.
- 60–70%
- Tickets resolved by agents
- < 5 min
- First response
- 0
- Refunds without approval
- Operations & compliance
Document intake
Every document read, checked and filed.
Contracts, forms, claims, IDs and statements extracted into your systems with 95%+ field accuracy, checked against policy, and only the exceptions left for a person.
- 95%+
- Field accuracy
- 70–90%
- Straight-through
- −80%
- Handling time
Before you ask
01Which models do you use?
Whichever fits the task and your data policy: Claude, GPT, Gemini or open models, often more than one. Model providers are configured with zero data retention.
02Where does the agent run?
In our managed cloud, your cloud account or on-premise. The choice is made in the design week based on your data and security needs.
03What happens if the agent gets something wrong?
Anything irreversible waits for approval, every action is logged and replayable, and the evaluation suite catches regressions before release.
Other services
Pair it with.
AI opportunity audit
Know exactly what to automate, and what it is worth.
We map how work actually moves through your teams and tools, score every candidate workflow on value, feasibility and risk, and hand you a costed roadmap. You decide what to build with real numbers in front of you.
2 to 3 weeksWorkflow automation & integrations
Your systems, finally talking to each other.
We connect the tools your business already runs on and automate the handoffs between them: forms to records, orders to fulfilment, tickets to teams, data synced both ways. Predictable steps run as automations; judgement calls go to an agent or a person.
1 to 4 weeks per workflowEnterprise AI rollout
AI across the company, with IT still in control.
We roll out personal agents and shared agents across departments: role templates, single sign-on, permissions, data-loss prevention, a governance console and the change management that gets people to actually use them. One platform instead of a dozen shadow tools.
Pilot in 6 weeks, rollout by department
Start with a 30-minute call.
We will tell you honestly whether this service fits, and what it would take.