You will get a custom AI agent that automates your workflows and uses your tools
Rising Talent
Project details
I build AI agents that do real work, not just chat. They use tools, call your APIs, read and update your data, and run multi step workflows on their own. I am a senior full stack and AI engineer (Python and FastAPI, Node, React and Next.js). I have shipped production agents, including an autonomous testing and CI agent and an agentic cost estimator with its own orchestrator, memory, and skills, and I connect LLMs from OpenAI or Claude cleanly into existing products. Every build ships deployed and documented, with a stabilization window so it keeps running after handoff. New on Upwork, so the Starter tier is a low risk first step: I build a working demo on your data before you pay anything. Tell me the workflow you want automated and I will show you exactly how I would build it and what it costs to run.
AI Algorithms
Large Language ModelAI Applications
AI-Generated Code, AIOps, Natural Language Generation, Natural Language UnderstandingAI Development Language
PythonAI Models
ChatGPT, GPT-4What's included
| Service Tiers |
Starter
$90
|
Standard
$450
|
Advanced
$1,500
|
|---|---|---|---|
| Delivery Time | 4 days | 10 days | 21 days |
Number of Revisions | 1 | 2 | 3 |
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
About Hitesh
Voice AI & AI Agent Engineer | LiveKit, RAG, Full-Stack
Hyderabad, India - 7:25 am local time
I don't prototype and disappear; I build things that run. A few I've shipped:
- Converscient: a real-time voice calling agent on LiveKit (speech in to LLM to speech out), built as independent microservices (FastAPI worker + React + MongoDB, Dockerized).
- An agentic vertical-AI cost estimator: an orchestrator coordinating task-specific agents over a typed knowledge base (FastAPI + Next.js + Postgres).
- DayPay: a multi-tenant payroll dApp with role-based access control and on-chain USDC payouts.
- Sentinel Guardian: an autonomous testing/CI agent that builds, then guards, a full test suite.
What I'm best at:
- Real-time voice AI: LiveKit, low-latency STT to LLM to TTS, barge-in, telephony.
- AI agents that take real actions through your tools and APIs (single- and multi-agent), not just chat.
- RAG chatbots grounded only in your own data (docs, sheets, sites), with citations.
- The full stack around it: React/Next.js, Node/Bun, Python/FastAPI, Postgres, Docker.
How I work: I scope the riskiest part first, ship a visible early win, and keep you updated every week. Model APIs (OpenAI/Claude) and hosting are billed to your own accounts, so you keep full control and visibility.
If you're building voice agents, agentic automation, or a chatbot on your own data, message me the one outcome you want and I'll tell you the fastest honest path to it.
Steps for completing your project
After purchasing the project, send requirements so Hitesh can start the project.
Delivery time starts when Hitesh receives requirements from you.
Hitesh works on your project following the steps below.
Revisions may occur after the delivery date.
Free demo and scope lock
I build a quick working demo on a sample of your data, then we lock the scope, the tools and APIs to connect, and what success looks like.
Build, integrate, and deploy
I implement the agent, the tool and API calls, and the workflow logic, deploy it to your cloud, and share progress in the workroom so you can follow along.