FAQ
Frequently asked questions
Direct answers to the questions we get most often. If yours is not here, email hello@studiobuildit.com.
Getting started
- How do I start an AI agent project with StudioBuildIt?
- Send us a message at studiobuildit.com/contact or book a 30-minute call at cal.com/studiobuildit/30min. You do not need a brief, a slide deck, or a detailed spec. A paragraph describing what you want to automate is enough to start. We respond within 24 hours on weekdays.
- How long does an AI agent build take?
- Most builds ship in 4 to 7 weeks. Voice agents take 5 to 7 days. WhatsApp support agents take 7 to 10 days. RAG customer support systems take 7 to 14 days. Lead qualification agents take 4 to 6 days. Custom vertical agents take 2 to 4 weeks.
- What does it cost to build an AI agent?
- Setup fees range from $1,500 for a simple workflow to $45,000 for a complex multi-agent system. Most single-agent builds fall between $4,000 and $15,000. We send a fixed-scope quote before any work starts. No hourly billing, no scope creep surprises.
- Do you work with clients outside India?
- Yes. Most of our clients are in the US, UK, UAE, Australia, and Canada. We are remote-first and have delivered 800+ projects without anyone needing to be in the same room. Time zone overlap is handled with async updates and scheduled calls.
How we build
- What stack do you build on?
- We build on Claude Sonnet 4.6 as the primary model, Mastra or LangGraph for orchestration, Qdrant for vector storage, Voyage AI for embeddings, Mem0 for persistent memory, and Langfuse for observability. For voice: Retell AI with Sarvam for Indian languages. For WhatsApp: WhatsApp Business API via AiSensy.
- Do AI agents make decisions without human approval?
- No. Every agent we build has explicit human-in-the-loop checkpoints for irreversible actions. The agent drafts, flags, or queues actions for a human to review before they execute. We document every checkpoint in the scope document before work starts.
- Can you integrate with our existing tools?
- Yes. Most builds include custom MCP server development to connect your internal systems to the agent. We have built MCP servers for Salesforce, HubSpot, Razorpay, Tally, Shiprocket, Zendesk, Intercom, Jira, Linear, and custom internal APIs.
- What model do you use?
- Claude Sonnet 4.6 for most builds. Claude Haiku 4.5 for high-volume, latency-sensitive tasks where cost matters. Sarvam Bulbul V3 for Indian-language voice (Hindi, Tamil, Telugu, Marathi, Bengali, Gujarati, and 5 more). We design model-agnostic: switching models is a config change, not a rebuild.
Pricing and ownership
- Do you charge for the model API usage?
- No. Model API keys live in your account and are billed directly to you. This keeps your data inside your perimeter and your costs transparent. We help you set up cost alerts and token budget monitoring as part of every build.
- What do we own after the project?
- Everything. You receive the full codebase in a private GitHub repo, CLAUDE.md architecture documentation, MCP server configs, a runbook, eval scripts, and Langfuse dashboard access. You can run, modify, and deploy the agent without us.
- What if the agent does not meet the acceptance criteria?
- We iterate until it does. The setup fee is not invoiced until you sign off on the live demo. We define acceptance criteria in writing before the demo, not during it. If we cannot deliver to your satisfaction, you owe nothing.
- Is there an ongoing retainer?
- Optional. The retainer covers monthly prompt audits, token budget monitoring, model updates when a new Claude version changes behavior, and up to 5 hours of changes per month. Pricing: $499 to $8,000 per month depending on complexity. We recommend it. One drifted agent mishandling calls costs more than the monthly retainer in a week.
Claude Skills and products
- What are Claude Skills suites?
- Claude Skills suites are pre-built, production-tested skill files for Claude Code, Cursor, Gemini CLI, and Codex CLI. Each suite covers a specific workflow: sales SDR, customer support, Indian SMB operations, content creation, DevOps, and more. One command installs the entire suite.
- Do Claude Skills work with Cursor?
- Yes. Every skill in every suite is cross-platform. The SKILL.md format is the same across Claude Code, Cursor, Gemini CLI, and Codex CLI. Place the files in your .cursor/skills/ folder and they activate automatically during chat.
- What is the refund policy on products?
- Full refund within 30 days if the skill does not work on your platform and we cannot fix it within 48 hours. Email hello@studiobuildit.com with your order ID and the specific issue. See studiobuildit.com/refund for the full policy.
Ready to ship something real?
Book a 30-minute build call. Walk away with a plan either way.