Frequently Asked Questions

Samcom Technologies offers AI voice agent development, AI chatbot and messaging solutions, VoIP development and support (Asterisk, FreeSWITCH, FreePBX, VitalPBX, 3CX), Flutter mobile app development, e-commerce development (Shopify, WooCommerce), web design and development (Laravel, React, Next.js), and staff augmentation for senior developers.

We are headquartered in Ahmedabad, India, and serve clients globally across the USA, UK, Australia, UAE, Canada, Singapore and Europe. Our developers work in your timezone with daily standups, Slack/Teams communication and transparent project management.

Samcom Technologies was founded in 2012 — over 14 years of delivering AI, VoIP, mobile and web projects globally. We have completed 500+ projects for 120+ clients across 30+ countries.

We start with a free consultation call to understand your requirements. We then provide a detailed proposal with scope, timeline and fixed or hourly pricing. Once agreed, we kick off with a project plan, assign a dedicated PM and communicate daily via your preferred channel.

Yes. We sign an NDA before any technical discussion, so you can describe the problem properly rather than in generalities. NDA and IP protection are standard on every engagement, including staff augmentation.

You do. Code is delivered to your repository and the provider accounts we set up — telephony, cloud, AI model providers — are opened in your name and billed to you directly. Recordings, transcripts and customer data stay in your systems and your accounts, not ours.

We primarily use LiveKit + GPT Realtime (gpt-realtime-2.1) + Deepgram for the lowest latency pipelines (sub-500ms). We also build on VAPI, Retell AI, Pipecat, Ultravox, Bland AI and VoiceFlow depending on budget and requirements.

On the pipeline we run in production, a call costs a fixed ₹1.24 the moment it connects plus ₹0.87 for every minute it runs — about ₹3.42 for a two-and-a-half minute enquiry call. Short calls therefore cost more per minute than long ones. Our voice AI cost page shows the full breakdown measured on real calls.

English plus six or more Indian languages, including Hindi, Marathi and Tamil. Language choice affects which speech and text-to-speech providers are the right fit, so we test the actual accent and vocabulary your callers use before committing to a stack.

Yes to both. Agents book into your calendar for appointment-led businesses — dental and medical practices, hotels, restaurants and salons — and transfer to a live person on request or when the conversation goes outside what the agent should handle.

A first working agent on a real number is typically live within a week. What takes longer is tuning: the prompt, the handover rules and the edge cases only appear once real callers are on it, so we plan for a period of listening to recordings and correcting.

Yes — this is one of our most common projects. We use the WhatsApp Business API with GPT-6 or Claude, LangChain for multi-turn memory, and bidirectional sync with Salesforce, HubSpot, Zoho or GoHighLevel.

Yes. We build on the Meta Cloud API directly rather than reselling a third-party layer, which is what makes custom template flows, media handling and webhook logic possible. We also run Pingly, our own WhatsApp CRM, if you would rather have the messaging desk ready-made.

Start with whichever channel your customers already use to reach you. Chat suits questions people are happy to type and re-read — order status, documents, pricing. Voice suits anything urgent, anything from a caller who will not install or type, and anything that needs a decision in one interaction.

Yes. VoIP troubleshooting is a core strength. We diagnose and fix one-way audio, SIP registration failures, choppy audio, dropped calls and SIP fraud — most critical issues are resolved the same business day using sngrep, Wireshark and platform logs.

Yes. We provide fresh installs on your chosen OS/cloud, configure SIP trunks, IVR, ring groups, voicemail, fail2ban hardening, and do zero-downtime migrations from legacy PBX systems.

Asterisk, FreeSWITCH, Kamailio, OpenSIPS, FusionPBX, FreePBX, VitalPBX, 3CX, Amazon Connect, and carrier platforms on PortaOne (PortaSwitch, PortaBilling and PortaSIP). We pick by what the workload needs — Kamailio or OpenSIPS in front for registration and load, Asterisk or FreeSWITCH behind it for media and application logic.

Yes. We set up direct routing so Teams users make and take PSTN calls through your own SIP trunks and your own PBX, rather than buying calling plans per seat.

Twilio, Plivo, Telnyx, Bandwidth, Sinch, Agora, Vonage, SignalWire and LiveKit. We also connect wholesale and local carriers directly where that works out cheaper than a CPaaS at your call volume.

Yes. We build WebRTC on LiveKit, mediasoup, Janus, Jitsi and Ant Media — SFU scaling, TURN/STUN, simulcast, recording, and browser-to-PSTN bridging so a web app and a phone number are the same conversation.

Flutter and Dart for most projects — one codebase for iOS and Android with BLoC, Riverpod or GetX and Firebase. We also build native in Swift and Kotlin, and maintain React Native codebases. We handle App Store and Play Store submission and the review responses that go with it.

Yes. Shopify Liquid and Shopify Plus theme work, custom apps on the Admin and Storefront APIs, Metafields and checkout extensions; WooCommerce for stores that need to stay on WordPress. We also connect stores to AI agents that follow up on abandoned carts and delivery questions by call or WhatsApp.

Yes. Salesforce, HubSpot, Zoho, GoHighLevel and Odoo are the ones we work with most. Integration usually means two-way sync rather than a one-off import, so a call, a chat or an order updates the record your team already works from.

You pick a role, we shortlist from our team, and you interview before committing. Rates run from $18 to $75 an hour depending on the specialism, billed monthly, with no recruitment fee. A developer can usually start within 48–72 hours of signing.

Yes. Developers work your hours, join your standups and use your tools — Slack, Teams, Jira or whatever you already run — rather than reporting into a separate agency process you cannot see.

A Vibe Coder builds with AI tooling — Claude Code, Codex, Cursor, Lovable, Replit and Kimi Code — with a senior engineer reviewing every line that ships. At $20–$30 an hour it suits internal tools, prototypes and admin panels where speed matters more than a bespoke architecture. For production systems that must be maintained for years, hire the specialist instead.

We offer three models: Fixed Price (clear scope and budget), Time & Materials (hourly for flexible or evolving projects), and Dedicated Team (monthly retainer for ongoing product work). All models include daily standups and transparent progress reporting.

For staff augmentation, we can have a developer start within 48–72 hours of signing the agreement. For new projects, we typically begin within 1 week of scope finalisation and contract signing.

Yes, under a separate support agreement rather than bundled vaguely into the build. That covers monitoring, patching, provider changes and an agreed response time for incidents. For telephony we run 24/7 monitoring and SLA-backed cover, because a PBX that fails at 2am fails for the whole business.

In your accounts. Recordings, transcripts, customer records and message history stay in the systems you own and are billed for. We work inside them for the engagement; we do not hold a copy afterwards, and there is no shared multi-tenant store holding your customers alongside somebody else’s.

Our privacy policy sets out the access, correction, deletion, portability and objection rights we honour, and the address to send a request to. Because your data stays in your accounts, you keep the ability to answer a subject request directly rather than waiting on us to extract it.