You will get custom n8n automation workflows to save your team hours
Project details
Most automations fail because they're built by someone who never asked how your team actually works. I do both sides: AI engineering and product design.
I build n8n workflow automations with AI built in — not just moving data between apps, but systems that read documents, make decisions, draft responses, and escalate to a human when they should.
What I build:
Document processing — extract, classify, and route data from invoices, contracts, forms
AI agents that handle multi-step tasks across your tools
Internal copilots that answer questions from your company's own documents
Lead, support, and reporting workflows with LLM steps
Human approval gates so nothing critical runs unchecked
You approve a workflow diagram before I build anything. Every project ships with error handling, retry logic, and documentation you can hand to someone else.
Background: AI Engineer & UX Designer at Invuric, previously Flashskool and Mintworxs. Engineering degree from NIT Hamirpur. Recent builds include an enterprise document intelligence system and an internal knowledge agent.
New to Upwork, not new to shipping production systems.
I build n8n workflow automations with AI built in — not just moving data between apps, but systems that read documents, make decisions, draft responses, and escalate to a human when they should.
What I build:
Document processing — extract, classify, and route data from invoices, contracts, forms
AI agents that handle multi-step tasks across your tools
Internal copilots that answer questions from your company's own documents
Lead, support, and reporting workflows with LLM steps
Human approval gates so nothing critical runs unchecked
You approve a workflow diagram before I build anything. Every project ships with error handling, retry logic, and documentation you can hand to someone else.
Background: AI Engineer & UX Designer at Invuric, previously Flashskool and Mintworxs. Engineering degree from NIT Hamirpur. Recent builds include an enterprise document intelligence system and an internal knowledge agent.
New to Upwork, not new to shipping production systems.
AI Algorithms
Convolutional Neural Network, Feedforward Neural Network, Large Language Model, Linear Discriminant Analysis, Long Short-Term Memory Network, Multimodal Large Language Model, Recurrent Neural Network, Regression Analysis, Transformer ModelAI Applications
AI Chatbot, AI Text-to-Image, AI-Generated Code, AIOps, Anomaly Detection, Conversational AI, Natural Language Understanding, Text RecognitionAI Development Language
PythonAI Tools
Adobe Firefly, Azure OpenAI, GitHub Copilot, Gradio, Hugging Face, Microsoft 365 Copilot, PyTorch, Replit, Streamlit, TensorFlowAI Models
BERT, ChatGPT, GPT-4, OpenAI Codex, WhisperWhat's included
| Service Tiers |
Starter
$95
|
Standard
$275
|
Advanced
$600
|
|---|---|---|---|
| Delivery Time | 3 days | 7 days | 14 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 |
Optional add-ons
You can add these on the next page.
Fast Delivery
+$40 - $200
Additional Revision
+$45
30-Day Post-Launch Support
+$150Frequently asked questions
About Rakshit
AI Automation Engineer | n8n, OpenAI, RAG Agents | Product-Minded
Palampur, India - 6:17 pm local time
I build AI systems businesses actually use: AI agents, internal copilots, chatbots, RAG document intelligence, and the dashboards teams work in daily. Not demos.
Most companies don't need another chatbot. They need AI that fixes a real workflow, fits how their team already works, and is simple enough that people adopt it. Combining hands-on AI engineering with product design experience is why my builds get used instead of abandoned after week two.
Recent work
Coates AI Workflow System — Enterprise document intelligence and process automation. Automated data extraction, human approval steps, error handling, and multi-step AI workflows built on n8n and LLM APIs.
Invuric Insights — Business intelligence agent giving teams instant access to company knowledge instead of manual document searching, wrapped in a dashboard built for daily use.
Aether — AI design agent combining LLM reasoning with UX principles to help users explore ideas and make design decisions faster.
How I approach a build
Before writing anything, I answer: Will people use this every day? Where should AI act, and where should a human stay in control? What happens at approvals and edge cases? I design the experience and build the intelligence behind it.
Stack: n8n, OpenAI & Claude APIs, LangChain, RAG and document intelligence, multi-agent systems, FastAPI, Supabase, API integrations, Figma, SaaS UX and dashboard design.
Services: AI automation, custom AI agents, internal copilots, AI chatbots, AI SaaS/MVP development, product and UX design.
Engineering background from NIT Hamirpur.
Send me your workflow or product idea and I'll reply with a short Loom walkthrough showing exactly how I'd build it.
Steps for completing your project
After purchasing the project, send requirements so Rakshit can start the project.
Delivery time starts when Rakshit receives requirements from you.
Rakshit works on your project following the steps below.
Revisions may occur after the delivery date.
Discovery & Process Mapping
I review your submitted process and tools, then send a short questionnaire or hop on a 30-minute call to clarify edge cases and confirm exact scope.
Workflow Design & Approval
You receive a diagram of the automation logic — triggers, AI steps, decision points, approval gates. I build nothing until you approve it.