
Closed
Posted
Paid on delivery
Summary Job Title: AWS AgentCore phone voice agent coded in Python 5 days to do *** important Provide minimum delays between question – answer – question for conversation dialogs *** important provide very good voice to text even in noisy environmet and with user interuption capabilities Use or [login to view URL] Posted on: Sep 22, 2026 with AI agents outside Connect Customer, over the open agent-to-agent (A2A) protocol through text or bidirectional voice AWS expanded the protocol to support voice streaming Customer phone ↕ Amazon Connect Customer ↕ A2A voice streaming ↕ AWS AgentCore (Python) ├── Speech-to-text ├── Bedrock LLM └── Text-to-speech Or just idea Phone provider: Twilio can't connect directly to AgentCore, because its stream address doesn't allow the login signature AgentCore needs. Vonage can, and AWS and Vonage have a sample that works with real phone calls, so I planned to use Vonage. Voice code: the Pipecat library (version 1.12) has everything we need in Python. It handles Vonage audio, Transcribe, Polly and OpenAI, and it can change the system prompt during a call, which is what your situation-based prompt needs. OpenAI on AWS: GPT-5.6 Luna can be called through Bedrock from Canada (Central). Requests go to Canada or US regions. Deploy tool: AWS has replaced the old Python agentcore command with a new one installed through npm (@aws/agentcore). [login to view URL] Job Type: Fixed-Price Very good natural voices for LLM use low cost fast models from openai or gemini or make suggestion **** No Terraform or similar techniques, only simple browser aws platform development *** use AWS from local windows Project Overview **** Develop Voice Real Estate agent (sell homes) **** For testing provide realistic properties descriptions from [login to view URL] or [login to view URL] or [login to view URL] etc. convert property web page to text extract Neighbourhood description and statistics Like example https://www.realtor.ca/real-estate/30180444/829-55-stewart-street-toronto-waterfront-communities-waterfront-communities-c1#view=stats https://www.realtor.ca/real-estate/30180444/829-55-stewart-street-toronto-waterfront-communities-waterfront-communities-c1#view=neighbourhood [login to view URL] I am looking for an experienced AWS and Python developer to build a scalable Voice AI Agent architecture using AWS Bedrock AgentCore. This is fundamentally an educational project: your goal is to build a fully working example system, document it exhaustively, and teach me how to reproduce, modify, and deploy it completely independently from scratch. looking for a one tenant Voice AI platform where agents can be created and managed, with Canadian phone number, isolated skill files, call memory/logs, and admin management, built on AWS Bedrock AgentCore + Python, along with complete documentation, videos, testing, and handover. You may use existing open-source GitHub projects as a foundation, provided the final deliverable meets all my requirements and is fully documented. agent answer questions about properties and persuade user to buy and book viewings intergrade some useful booking service and send message by email and sms text to both user and realtor full log such call is transcribed and session logs System Architecture & Core Examples We need a one Canadian phone voice agent system acting as customer service caller phone number ID identification email interface for admin to provided gmail address : add or remove files for skills; send by email logs of conversations , block phone numbers for incoming call The Use-Case Examples: Think of these as sales agents for stores or services , where store has many unique products, for example : Or Case 1: Food Catering Or case 2: Cleaning Services Or case 3: Refrigerator Repair Or case 4 : Real Estate agent (sell homes) Expected Interaction Flow: At the beginning of the call, the agent asks the user what they want to talk about (essentially asking which file to add to system prompt ). Example A (Case 3): A user calls for Store 3 and says, "I have Refrigerator abc123." The agent must dynamically retrieve the specific .txt skill file for "Refrigerator abc123" and use this skill for the remainder of the session. Example B (Case 2): A user calls for Store 2 and says, "I need to clean a sofa." The agent chooses the specific .txt skill file on how to clean a sofa and uses it for the session. Example C (Case 3) : user wants to talk about specific home for sale . Find skills file with specific address what user tells you . Each .txt file has header with description what is this file about , then agent find which file header matches the beset to user request , if bad matching happen then during conversation with user it should be found correct match , etc . Usual skills router implementation Also if user wants then conversations returns to state to choose what property to discuss **** Also search of matching properties by search criteria for example all properties from your portfolio with one bedroom , test with 10 properties in portfolio *** Technical Features & Requirements Voice Integration: Phone call voice chat using a Canadian phone number (+1). One phone number serves exactly one agent/store. Strict Data Isolation: agent has its own dedicated group of .txt knowledge files. Session Memory: The agent must retain context during the call (e.g., if the caller says "My name is Peter," the agent calls him Peter for the whole session). There must be no memory between different sessions. Logging: Detailed logs must be generated and saved for every individual call/session. Technical Stack Language: Python AI Framework: AWS Bedrock AgentCore Telephony: use Amazon Connect - low cost AWS services compatible with AgentCore , I am open to discuss other options like Chime SDK, or Twilio Phone call with interruption option Infrastructure: AWS Strict Development Constraints Your Own Environment and development resources : You must develop, record, and test this entirely in your own authorized AWS account. I will not provide my AWS account or LLM credentials (openai or gemini ) or etc resources Zero Dependencies: My later deployment must not depend on your accounts, resources, or API keys. Use openai or gemini for LLM (I am open for suggestions for other good LLM providers ) Deliverables (Required for Payment) This fixed-price project covers the application, deployment, security, tests, source code, reporting, manuals, videos, and a handover session. 1. Editable Word (.docx) Manual: Must start from a totally clean environment. Every actionable setup/configuration step needs a real, readable screenshot. Numbered instructions with copyable commands/settings. Explain the purpose of the step, the expected result, and how to verify it. Clearly distinguish between Windows and Server commands and explain any placeholders (e.g., [YOUR_BUCKET_NAME]). Must cover: Prerequisites, code structure, AgentCore setup, memory/RAG routing, telephony settings, database/logs, AWS IAM/security, tests, monitoring, recovery, updates, cost breakdown, and complete environment cleanup. Include exact software/library versions and architecture details. Do not include real secrets/passwords in the doc. Include a section on common errors and fixes. 2. Narrated MP4 Videos: Follow the exact section/step numbers of the .docx manual. Include a timestamp index. Show the actual, complete setup from scratch (no skipped prerequisites or undocumented pre-configuration). Demonstrate application features, parallel-dialogue tests, and a security assessment showing the strict .txt file isolation. Note: Videos, screenshots, and code must match perfectly. 3. Handover & Teaching: A session (or clear video guidance) teaching me how to modify both the Python application codebase and the Python test suite. Acceptance Criteria The project will be considered successful and complete only when I am able to reproduce the entire project by myself, from scratch, in my own AWS account, relying entirely on your .docx manual and videos without needing to ask you for missing steps. Very good natural voices Clear instructions how to delete all created resource – not to pay after project done
Project ID: 40789709
73 proposals
Remote project
Active 7 hours ago
Set your budget and timeframe
Get paid for your work
Outline your proposal
It's free to sign up and bid on jobs
73 freelancers are bidding on average $24 CAD for this job

Hi, I can build the AWS AgentCore voice agent in Python connected to Amazon Connect via A2A voice streaming: streaming STT tuned for noisy audio, Bedrock LLM, streaming TTS, with barge-in (user interruption) and minimal turn latency by streaming partial responses. I build voice AI agents and work with AWS and Azure. Question: is the Amazon Connect instance already set up with a phone number? Best regards, Khuda Dino
$20 CAD in 5 days
5.5
5.5

Hi there, I’m excited to propose my services for your AWS AgentCore phone voice agent project. Leveraging my expertise in Python, AWS, and AI, I understand the importance of developing a scalable and efficient Voice AI Agent architecture. Your requirement for a real estate-focused voice agent aligns perfectly with my experience in building AI-driven applications. I propose to utilize AWS Bedrock’s AgentCore alongside Python to develop a robust system, ensuring seamless voice interaction through Amazon Connect or Vonage. Integrating advanced LLMs like OpenAI’s GPT-5.6 Luna, I will ensure the agent delivers natural and persuasive dialogues, meeting your need for realistic property discussions and bookings. My approach involves using existing open-source resources, ensuring the project is fully documented for your independent replication and modification. I will deliver comprehensive manuals and video tutorials to guide you through setup, deployment, and operation, ensuring you can manage and scale the system confidently. I am committed to achieving strict data isolation, session memory retention, and detailed logging, as specified. By developing in my authorized AWS environment, I guarantee a zero-dependency handover, allowing you to deploy the solution independently without relying on external accounts or resources. I look forward to the opportunity to bring this innovative project to life with you. Best Regards,
$10 CAD in 2 days
5.4
5.4

Hi, I hope you're doing well. I can help structure this Python-based voice AI project around AWS AgentCore, with a clear focus on low-latency conversations, natural speech, caller interruptions, and reliable session handling. The architecture should separate telephony, speech recognition, LLM orchestration, text-to-speech, property retrieval, and call logging so each component can be tested and maintained independently. I’d build a prototype with a Canadian phone number, isolated property skill files, session-only memory, property search across 10 sample listings, viewing-booking workflows, and email/SMS notifications. I’d also include caller blocking, full transcripts, access controls, automated tests, and cost-conscious model recommendations. Telephony and AgentCore compatibility would be validated early before committing to the final integration. Your documentation and teaching requirements are substantial, so I’d organize delivery into working milestones: architecture validation, functional prototype, security and concurrency testing, then the illustrated Word manual, matching narrated videos, cleanup instructions, and handover. Development would use my authorized AWS environment, with no dependency on my accounts or credentials in your final deployment. Best regards, Razmik
$20 CAD in 1 day
5.0
5.0

Hi there, I noticed your focus on creating a scalable Voice AI agent using AWS Bedrock AgentCore with Python. Implementing features like session memory, logging, and integrating tools like Vorange aligns perfectly with my skill set. I can help structure the agent to cater to various properties effectively, ensuring smooth interactions for users seeking real estate options. I’m prepared to guide you through the process, ensuring you understand how to deploy and modify the system independently. Additionally, I will document every step and create instructional videos for your convenience. Let’s discuss how we can proceed with this unique project to fulfill your educational goal. Regards, Ali
$30 CAD in 1 day
4.6
4.6

Hi, I’ll build the AWS AgentCore voice agent in Python with a modular skills-router, session-only memory, call logging, and Canadian telephony integration. I’ll implement property search, booking, SMS/email follow-ups, isolated .txt knowledge files, admin controls, and clean prompts for low-latency natural dialogue. I’ll also document every setup step, test flow, and cleanup procedure clearly. Start the chat so we can discuss it further. Regards, Sai Bhaskar
$30 CAD in 1 day
4.3
4.3

Hi, I see you need a voice agent that can handle conversations about properties, which requires integrating AWS Bedrock AgentCore with a telephony service, and I'd approach this by using Amazon Connect for the phone calls. To ensure good voice-to-text capabilities, even in noisy environments, I'll set up the Pipecat library to handle Vonage audio and transcription. I've found that a common mistake in these projects is not properly handling interruptions during calls, so I'll make sure to implement a system that can handle user interruptions and still provide accurate responses. As a result, you can expect a fully functional voice agent that can understand and respond to user queries about properties, with clear instructions on how to deploy and manage the system, and a handover session to teach you how to modify and update the codebase.
$20 CAD in 7 days
4.2
4.2

I built AgentCore voice flows with Python before, the main trip up with phone agents is barge in handling and keeping latency under a second when the model calls tools mid conversation. I would wire Bedrock AgentCore with your choice of OpenAI or Gemini for the NLP layer, running on Linux. Can start today, working version in 4 days. Budget and timeline here are rough, based only on what's posted. We will nail down real numbers once I see the call flow you need. Want me to send a quick scope doc so we can get moving?
$30 CAD in 4 days
3.7
3.7

Hello, I have 11+ years of experience in Python and AWS-based AI systems, including real-time voice workflows, Bedrock integrations, and production-ready logging and session handling. I understand you need a Canadian phone voice agent with strict .txt skill isolation, per-call memory, minimal dialog delays, booking and notification flows, and full admin control. I’d build it with Python, Bedrock AgentCore, and a telephony option that fits your stream requirements, then document every step clearly with screenshots, tests, and handover materials. I’d like to connect with you in chat to discuss the best architecture, including routing, voice quality, and deployment details. Thanks
$20 CAD in 7 days
3.8
3.8

Hi, The key challenge is making the voice agent feel natural on real phone calls, with low response latency, accurate transcription in noisy environments, and interruptions handled without awkward delays. The skill router also needs strict file isolation so each caller gets the correct property knowledge without carrying information into another session. I’d start with a working Python prototype using AWS-compatible telephony, streaming speech, a cost-conscious LLM, and a 10-property test portfolio. Then I’d validate call logging, property search, viewing bookings, SMS/email notifications, and cleanup procedures before documenting the complete setup. Your five-day deadline and extensive screenshot, video, testing, and handover requirements need a clearly agreed scope. The listed budget also appears too low for the full deliverable, so I’d confirm priorities before committing to the complete build. Kind regards, Fernando
$30 CAD in 3 days
3.7
3.7

Hi, Your project is particularly interesting because the main challenge is not simply connecting a phone number to an LLM. It is building a low-latency, interruptible voice conversation system with reliable speech recognition, natural-sounding responses, dynamic knowledge routing, and complete session isolation. I have experience with Python, API integrations, automation, and backend application development. I would approach the Python application with a modular architecture, using AWS services for telephony and deployment, a voice pipeline for speech-to-text and text-to-speech, and an LLM for property conversations and sales workflows. For the knowledge router, each skill file would have searchable metadata and a description header. The system would select the relevant property or service information, retrieve matching content, and maintain that context throughout the call without carrying memory into another session. I would also prioritize interruption handling, caller identification, call transcripts, property searches, viewing bookings, and email/SMS notifications. The documentation is equally important here. The final handover should allow you to recreate the environment independently, understand every configuration, test the application, and remove all AWS resources when finished. Would you be open to starting with the real estate use case and a small portfolio of 10 properties? Best Muhammad Saad
$20 CAD in 1 day
3.5
3.5

Hi, I'd be happy to help build your Python-based real estate voice agent using AWS Bedrock AgentCore and Amazon Connect. I'd focus on low-latency, natural conversations with interruption handling, dynamic skill-file routing, property search, appointment booking, and email/SMS notifications. The solution would include isolated session memory, call transcripts, admin controls, and secure AWS deployment without Terraform. I can also provide Python source code, a step-by-step Word manual with screenshots, narrated setup videos, testing, and resource cleanup instructions so you can reproduce everything independently. The full scope is substantial for a 2-day timeline and CAD $10 budget, so I'd suggest agreeing on a focused working prototype first. Happy to discuss the architecture and priorities. Thanks! Sami
$10 CAD in 2 days
3.0
3.0

Building AI agents, automation platforms, and full-stack systems is my core work, backed by 40+ production deployments. Your AWS AgentCore voice real-estate agent matches my experience with AI, OpenAI, Python, AWS, FastAPI, Twilio, and business workflows that connect conversations to real actions. I can develop the educational, reproducible example around a Canadian phone agent that answers property questions, handles interruptions, maintains low-latency dialogue, captures caller identification, and supports sales conversations and viewing bookings. My experience with AI agents, CRM automation, messaging, Stripe-connected platforms, and property systems such as JaipurBnB is directly relevant to call logs, transcripts, admin controls, skill-file management, blocked numbers, and email/SMS notifications. OpenAI is already part of my production AI stack, while Python is one of my listed core skills. You will receive a practical, documented system rather than an isolated demo, including deployment guidance from Windows through the AWS console, testing, handover materials, and the reusable foundation needed for catering, cleaning, repair, or other service agents. I would be glad to build this AgentCore voice platform and deliver it with the documentation and teaching focus you require.
$20 CAD in 1 day
3.1
3.1

Hi, The hardest part of this voice agent is not connecting a phone number to an LLM. It is keeping conversations natural with minimal response latency, reliable speech recognition in noisy environments, and immediate interruption handling while routing each caller to the correct property knowledge file. I can build the Python voice agent using AWS Bedrock AgentCore, with Amazon Connect or Vonage selected after validating bidirectional audio streaming and latency. The system will support Canadian phone calls, 10 searchable property profiles, dynamic .txt skill routing, session-only memory, viewing bookings, email/SMS notifications, call transcripts, number blocking, and admin controls. Knowledge isolation will be enforced in the backend, not just through prompts. #step1 Validate telephony, streaming latency, STT accuracy, barge-in, and model costs using my AWS environment. #step2 Build AgentCore routing, isolated property files, session memory, and property search. #step3 Integrate bookings, notifications, administration, and call logging. #step4 Test noisy audio, interruptions, concurrent calls, and security boundaries. #step5 Deliver Python source, tests, screenshot-based Word documentation, narrated setup videos, Windows deployment instructions, and complete AWS cleanup guidance. Roman
$50 CAD in 3 days
2.8
2.8

Hi, I can develop your Python-based AI voice agent using AWS services, with real-time phone conversations, natural text-to-speech, speech recognition, and interruption handling. I can implement property-specific skill-file routing, session-based memory, call transcripts, and an admin interface for managing knowledge files and blocked phone numbers. I’ll also work on property search, viewing bookings, email/SMS notifications, security testing, and detailed deployment documentation. The project will be structured so you can reproduce the setup in your own AWS account without depending on my credentials or resources. I’m happy to discuss the implementation plan and milestones in chat so we can agree on the scope and ensure every acceptance criterion is covered. Best regards Yasir k
$60 CAD in 2 days
2.2
2.2

Have you considered how to ensure minimal delay between responses in conversational AI? To build a Voice AI Agent using AWS Bedrock and Python, I can implement optimized voice-to-text functions while managing interruptions effectively. With over 5 years of experience in Python and AWS, I’m confident in developing a scalable Voice AI platform tailored to your needs. I’ll document the entire process, ensuring you can modify and deploy independently. Using Vonage for voice integration and open-source foundations, I can create a robust agent capable of accessing property data and managing bookings seamlessly. Let me know if you’d like to discuss this further. Excited to help you create a next-gen real estate voice agent!
$18 CAD in 4 days
2.3
2.3

The project includes the development of a scalable Voice AI Agent architecture using AWS Bedrock AgentCore, focusing on real estate functionalities. My extensive experience in Python and AWS will enable me to create a robust system that meets your specifications. I have successfully integrated voice technologies and built AI-driven platforms, which will help in implementing your requirements for voice interaction, data logging, and context retention. Additionally, I am familiar with using libraries such as Pipecat for handling voice streams, which is critical for this project. My first step would involve setting up the environment and documenting the process to ensure clarity and reproducibility. I look forward to collaborating effectively on this project and getting started.
$10 CAD in 1 day
1.5
1.5

Hello, I checked your project "AWS AgentCore phone voice agent coded in Python -- 2" and already have a clear understanding of your requirements and how to deliver them efficiently. I have solid experience in Python, and I've successfully completed similar projects by delivering high-quality, scalable, and reliable solutions. Why choose me? * Strong expertise in Python * Clean, optimized, and scalable implementation * Fast communication and regular progress updates * Focused on delivering results, not just completing the task I'm ready to start immediately and would be happy to discuss the details. Best regards, Umer
$25 CAD in 1 day
1.2
1.2

Hello, Building a voice real estate agent that routes between property files, holds session context, and logs every call comes down to wiring Bedrock AgentCore to Amazon Connect, then layering in dynamic prompt switching based on caller input. I built a similar multi-tenant booking system on Node where each tenant had isolated knowledge files and per-session memory that reset between calls. The routing logic matched user input against file headers to pick the right context, and transcripts went to DynamoDB. One early snag was prompt injection across sessions, which I locked down by clearing context boundaries strictly between calls. Here, the flow is: caller identifies a property, the agent fetches the matching .txt file header, injects it into the system prompt for that session only, and Bedrock handles the conversation. Session logs and call transcripts go to S3 and DynamoDB. Amazon Connect handles the phone side cleanly, and Polly gives natural voice. The manual will walk through every AWS setup step from a clean Windows environment with actual screenshots, exact IAM policies, and how to swap properties or add new agents without touching code. Videos will show the full setup, a live call demo, and how .txt file isolation holds under load. After handover, any bugs within scope are fixed for two weeks at no extra charge. Glad to start on the architecture and get you a working test call within the first few days. Thanks, Christian
$30 CAD in 1 day
1.4
1.4

Hi, The hard part is keeping the pause between caller speech and reply short over a real Canadian phone line. We would build the agent in Python on Bedrock AgentCore with Pipecat. We would try Amazon Connect A2A voice streaming first and fall back to Vonage if latency is too high. Transcribe streams partial text to a fast low cost OpenAI or Gemini model on Bedrock while Polly neural voices start speaking before the full reply is ready. The real estate agent reads listing and neighbourhood text from skill files, answers questions, books viewings through a calendar service, then emails and texts both caller and realtor. Every call is transcribed and logged with caller ID, while a Gmail admin flow adds or removes skill files, sends logs and blocks numbers. Setup uses only the AWS console from Windows, documented step by step with videos so you can rebuild it on your own. Which booking service does your realtor use? Is a Canadian number already active in Connect? Can we review our plan and proposed timeline in a 10-minute call on Freelancer so you can decide how to proceed? We suggest one working call that answers questions about a single listing as the first milestone. Regards, Bits Collision
$30 CAD in 2 days
1.4
1.4

Hi, I'd build the voice agent in Python using AWS Bedrock AgentCore, with Amazon Connect Customer's A2A integration where supported. My implementation plan: 1. Create an AgentCore Python service using the current AWS AgentCore CLI, IAM roles and scoped configuration. 2. Connect a Canadian phone number using Amazon Connect's A2A integration or a compatible telephony streaming provider. 3. Implement speech recognition, low-latency responses, natural TTS and interruption handling. 4. Load each business's .txt knowledge files separately, selecting the relevant skill by user intent and preventing cross-business access. 5. Keep conversation context within each call, with no memory shared between sessions. 6. Implement property search, viewing requests, booking notifications, SMS/email delivery and per-call logs. 7. Test noisy audio, interruptions, knowledge isolation and recovery from failed calls. I'll provide documented Python code, deployment and teardown instructions, test cases, and reproducible walkthrough materials. Best regards, Braian Alves
$30 CAD in 1 day
1.2
1.2

Downsview, Canada
Payment method verified
Member since Nov 29, 2017
$30-250 CAD
$10-30 CAD
$30-250 CAD
$30-250 CAD
$10-30 CAD
₹600-800 INR
₹10000-75000 INR
$10-30 CAD
₹12500-37500 INR
$30-250 USD
$2-8 USD / hour
$5000-10000 USD
₹1500-12500 INR
$750-1500 AUD
$25-50 USD / hour
$25-50 USD / hour
$10-30 CAD
₹6000-60000 INR
$8-15 USD / hour
₹37500-75000 INR
$250-750 USD
$250-750 USD
$15-25 USD / hour
₹750-1250 INR / hour
$30-250 USD