AI Automation Engineer
Build intelligent workflows that save your team hours every week
Problem: Your team repeats the same manual steps every week: copying data between tools, chasing updates, writing first drafts. Off-the-shelf automations break as soon as the process has an exception.
- Custom AI agents (n8n, Make.com, LangChain)
- Workflow automation design + implementation
- API integrations between your tools
- Zapier / Make migration to n8n
- Prompt engineering & LLM integration
- Weekly progress reports
- Timeline
- First deliverable in 2 weeks
- Best for
- Startups and scale-ups who want to automate operations fast
How it works
- 01
Discovery call
A 30-minute call about the processes that cost your team the most time, the tools involved and what a good result looks like.
- 02
Map and prioritise
I map the workflows worth automating first and agree the scope for the first month in writing.
- 03
Build in short cycles
Each workflow or agent is built, tested on real examples from your team and handed over, with a written update every week.
- 04
Run and improve
Once a workflow is live, I monitor failures, handle edge cases and move to the next item on the list.
What I need from you
- One person on your side who owns the process and can answer questions
- Access to the tools involved (or a test workspace), with API keys created under your accounts
- A handful of real examples of the work you want automated, including the awkward cases
- Agreement on where the automations run (your n8n instance or one set up for you)
Questions
Which tools do you work with?
Mostly n8n, plus Make.com, Zapier, LangChain and direct API integrations. I use what fits your stack; moving Zapier or Make workflows to n8n is a common part of the work.
How does the retainer work?
A monthly retainer covers an agreed amount of work, billed monthly in advance; hourly work is available for smaller requests. Larger one-off builds are split into milestones agreed in the proposal.
Who owns the workflows and code?
You do. Everything built for you, including workflows, code, prompts and documentation, belongs to you once it is paid for, and it runs under your own accounts.
What happens when an automation fails?
Failures are logged and alerted, and while the retainer runs I fix issues and adjust workflows as your process changes.
Will you sign an NDA?
Yes. I am happy to sign your NDA before we discuss details.
How do we communicate?
I am based in India (IST) and keep 3–4 hours of overlap with US or European working hours, with a written update every week.
Related work
- Case studyContent pipeline automation for a digital marketing agencyAn agency that writes for several clients had a production line held together by senior writers' time. Every article started with hours of research and d…
- Case studyReal-time inventory sync for a multi-channel e-commerce retailerA retailer selling the same stock on three channels kept selling items it no longer had. The cause was not demand or a bad supplier: it was a reconciliation…
- Projectn8n Content MachineAn n8n workflow that turns one topic from a Google Sheet into 12 platform-specific drafts in Google Docs with a single Gemini call. Drafts only.
- ProjectAI Inbox AgentAn inbox agent that classifies email, summarises it and drafts replies from your past answers, with human approval. A Python CLI and an n8n + Slack version.