You will get a real-time AI voice agent that makes and receives phone calls
Rising Talent
Project details
A voice agent should feel like a person who picks up on the first ring, not a phone tree. I build real-time AI agents that answer and place actual phone calls, understand the caller, take action, and hand off to a human when it matters.
Built on LiveKit and WebRTC with a streaming STT to LLM to TTS pipeline, my agents reply under a second and connect to real phone numbers over SIP (Twilio, Telnyx, or your own PBX). Inbound support lines, outbound callers, AI receptionists, appointment booking, all with call recording and transcripts.
4+ years shipping AI in production. A Shopify voice agent handling support and sales across 1,881 products. A HIPAA-compliant deployment on a private GPU cluster.
What you walk away with:
• A voice agent that makes and receives calls in real time, sub-second latency
• SIP telephony wired to your numbers, with recording and transcripts
• Tool calling: booking, CRM lookups, live transfer to a human
• Clean code you own, deployed and documented
Stack: LiveKit, WebRTC, SIP, Twilio, Deepgram, OpenAI, ElevenLabs, Python, FastAPI.
Send me the call flow you want handled and I will show you how I would build it.
Built on LiveKit and WebRTC with a streaming STT to LLM to TTS pipeline, my agents reply under a second and connect to real phone numbers over SIP (Twilio, Telnyx, or your own PBX). Inbound support lines, outbound callers, AI receptionists, appointment booking, all with call recording and transcripts.
4+ years shipping AI in production. A Shopify voice agent handling support and sales across 1,881 products. A HIPAA-compliant deployment on a private GPU cluster.
What you walk away with:
• A voice agent that makes and receives calls in real time, sub-second latency
• SIP telephony wired to your numbers, with recording and transcripts
• Tool calling: booking, CRM lookups, live transfer to a human
• Clean code you own, deployed and documented
Stack: LiveKit, WebRTC, SIP, Twilio, Deepgram, OpenAI, ElevenLabs, Python, FastAPI.
Send me the call flow you want handled and I will show you how I would build it.
AI Algorithms
Autoencoder, Convolutional Neural Network, CycleGAN, Generative Adversarial Network, Large Language Model, Multimodal Large Language Model, Recurrent Neural Network, Transformer Model, Variational Autoencoder, YOLOAI Applications
AI Content Creation, AI Text-to-Speech, Automatic Speech Recognition, Conversational AI, Natural Language Understanding, Sequence Modeling, Speech Synthesis, Text RecognitionAI Models
ChatGPT, GPT-4, OpenAI Codex, WhisperWhat's included
| Service Tiers |
Starter
$499
|
Standard
$1,499
|
Advanced
$3,499
|
|---|---|---|---|
| Delivery Time | 6 days | 12 days | 24 days |
Number of Revisions | 2 | 4 | 6 |
AI Model Integration | |||
Batch Normalization | - | ||
Database Integration | |||
Detailed Code Comments | |||
Image Upscaling | - | ||
MLOps | - | ||
Model Deployment | - | ||
Model Documentation | |||
Model Monitoring | - | - | - |
Model Testing & Optimization | - | ||
Model Tuning | - | - | |
Natural Language Processing | |||
NLP Tokenization | |||
Pre-Training | - | - | |
Prompt Engineering | |||
Setup File | |||
Source Code |
Frequently asked questions
2 reviews
(2)
(0)
(0)
(0)
(0)
This project doesn't have any reviews.
ES
Ellis Otto S.
Jul 4, 2026
Fix Poor Retrieval Accuracy in Our RAG-Based AI Agent (FastAPI + OpenAI)
Really solid work. Our AI agent was missing obvious answers and hallucinating, and Abid quickly got to the root of it and fixed the retrieval quality within the week. He clearly knows RAG systems inside out and explained everything along the way, so we understood exactly what was changed and why. Great communication throughout and I'd happily work with him again.
GH
George H.
Jun 22, 2026
WebRTC Realtime Engineer Fix & Optimize Low-Latency Audio/Video Connection (Fixed Project)
Abid really knows his stuff with WebRTC. We had connection drops that kept coming back and he found the actual cause and fixed it, then got our own TURN server running too. Easy to talk to and kept me updated the whole time. Would definitely hire him again.
About Abid
AI & Full-Stack Developer | React, Next.js, Node.js, Python & WebRTC
100%
Job Success
Kingston upon Hull, United Kingdom - 9:22 am local time
Core Expertise
𝐅𝐮𝐥𝐥-𝐒𝐭𝐚𝐜𝐤 𝐖𝐞𝐛 & 𝐒𝐚𝐚𝐒 𝐃𝐞𝐯𝐞𝐥𝐨𝐩𝐦𝐞𝐧𝐭:
I design and deliver robust web platforms using 𝐑𝐞𝐚𝐜𝐭, 𝐍𝐞𝐱𝐭.𝐣𝐬, 𝐍𝐨𝐝𝐞.𝐣𝐬, 𝐚𝐧𝐝 𝐄𝐱𝐩𝐫𝐞𝐬𝐬, with 𝐌𝐨𝐧𝐠𝐨𝐃𝐁 𝐨𝐫 𝐏𝐨𝐬𝐭𝐠𝐫𝐞𝐒𝐐𝐋 as the data layer. I build secure, high-performance frontend and reliable backend APIs that scale effectively.
𝐀𝐈 & 𝐌𝐚𝐜𝐡𝐢𝐧𝐞 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠:
I integrate advanced AI capabilities into products, including 𝐜𝐨𝐦𝐩𝐮𝐭𝐞𝐫 𝐯𝐢𝐬𝐢𝐨𝐧, 𝐎𝐂𝐑, 𝐍𝐋𝐏, 𝐚𝐧𝐝 𝐥𝐚𝐫𝐠𝐞 𝐥𝐚𝐧𝐠𝐮𝐚𝐠𝐞 𝐦𝐨𝐝𝐞𝐥𝐬 (𝐋𝐋𝐌𝐬). Notably, I built an AI diagram scanner using 𝐑𝐅-𝐃𝐄𝐓𝐑 𝐚𝐧𝐝 𝐏𝐚𝐝𝐝𝐥𝐞𝐎𝐂𝐑 𝐰𝐢𝐭𝐡 𝐒𝐀𝐌𝟐 𝐬𝐞𝐠𝐦𝐞𝐧𝐭𝐚𝐭𝐢𝐨𝐧, achieving 𝟗𝟕% 𝐚𝐜𝐜𝐮𝐫𝐚𝐜𝐲. I also fine-tune LLMs for chatbots and develop RAG-based recommendation systems using Pinecone and AWS Bedrock.
𝐑𝐞𝐚𝐥-𝐓𝐢𝐦𝐞 𝐂𝐨𝐦𝐦𝐮𝐧𝐢𝐜𝐚𝐭𝐢𝐨𝐧 & 𝐒𝐭𝐫𝐞𝐚𝐦𝐢𝐧𝐠:
I develop interactive, low-latency applications like video conferencing and virtual classrooms using 𝐖𝐞𝐛𝐑𝐓𝐂, 𝐋𝐢𝐯𝐞𝐊𝐢𝐭, 𝐉𝐚𝐧𝐮𝐬, 𝐅𝐅𝐦𝐩𝐞𝐠, 𝐚𝐧𝐝 𝐆𝐒𝐭𝐫𝐞𝐚𝐦𝐞𝐫 for smooth, real-time media streaming experiences.
𝐂𝐥𝐨𝐮𝐝 𝐀𝐫𝐜𝐡𝐢𝐭𝐞𝐜𝐭𝐮𝐫𝐞 & 𝐃𝐞𝐯𝐎𝐩𝐬:
I design and automate cloud infrastructure on 𝐀𝐖𝐒, 𝐆𝐂𝐏, and Azure using 𝐃𝐨𝐜𝐤𝐞𝐫, 𝐓𝐞𝐫𝐫𝐚𝐟𝐨𝐫𝐦, 𝐚𝐧𝐝 𝐂𝐈/𝐂𝐃 𝐩𝐢𝐩𝐞𝐥𝐢𝐧𝐞𝐬. My focus is on secure, scalable, and fault-tolerant deployments supported by best DevOps practices.
𝐇𝐨𝐰 𝐈 𝐖𝐨𝐫𝐤:
I handle the full development cycle, from initial architecture and prototyping through implementation, testing, and deployment. I focus on clear communication, realistic timelines, and writing maintainable, well-documented code so your team can work with it easily in the future.
If you’re looking for a developer who bridges 𝐟𝐮𝐥𝐥-𝐬𝐭𝐚𝐜𝐤, 𝐀𝐈, 𝐚𝐧𝐝 𝐫𝐞𝐚𝐥-𝐭𝐢𝐦𝐞 𝐬𝐲𝐬𝐭𝐞𝐦𝐬, I’d be glad to collaborate. Share your project details, and I’ll propose a clear, tailored plan with the right architecture, milestones, and tech stack to achieve your goals efficiently.
Steps for completing your project
After purchasing the project, send requirements so Abid can start the project.
Delivery time starts when Abid receives requirements from you.
Abid works on your project following the steps below.
Revisions may occur after the delivery date.
1.
Discovery and call flow We get on a call. I map what the agent should say, do, and when it should transfer to a human.
2.
Telephony and architecture I set up your phone numbers, SIP trunk, and the LiveKit + WebRTC real-time pipeline.