AI Automation Engineer in India
AI automation connects language models with business tools so a workflow can interpret unstructured information and take a defined action. I’m Anurag Srivastav, an AI automation engineer in New Delhi, India. My portfolio includes lead-processing pipelines, multilingual feedback routing and AI deployment tools.
By Anurag Srivastav · New Delhi, India · Open to remote projects
Discuss your projectWhere AI automation can help
A useful starting point is a repetitive task with a clear input and outcome: categorizing incoming feedback, enriching a lead record, preparing a report or routing a request. Language models can help interpret text; APIs and ordinary code handle the surrounding data movement and business rules.
Choose one workflow before automating an entire department. Document who owns it, what a correct result looks like, which systems it touches and when a person should review the output. This makes the scope easier to evaluate and prevents an impressive demo from becoming a difficult production dependency.
What we can scope together
Depending on the workflow, the project can include an input trigger, Python or n8n processing, model integration, database updates and notifications. A useful handover includes the workflow definition, configuration instructions and an explanation of how to handle failures.
For consequential actions, define an approval step. For repeatable actions, define a stable identifier so retries do not create duplicate records or notifications. Keep the model’s task narrow, and make exceptions visible to the people responsible for the process.
How to evaluate a pilot
Compare the pilot with the existing manual process using the same sample of tasks. Record correct outcomes, exceptions requiring review, processing time and cost per completed task. Agree on acceptance criteria before expanding the workflow.
A realistic project brief includes sample inputs with sensitive details removed, the tools already in use, typical volume, the expected output and any approval requirements. These details are more useful for estimating the work than a request to simply add AI.
Related work from my portfolio
AI Outreach Automation Pipeline
Fully autonomous outreach system that scrapes LinkedIn profiles, enriches lead data, generates personalized GPT emails and sends 200+ emails daily.
Tools: n8n, GPT-4, Supabase, SMTP, Twilio
Explore the portfolio projectsBMRC Hospital AI Feedback Agent
Enterprise AI workflow that processes multilingual patient feedback, runs sentiment analysis, routes tickets automatically and sends Gmail notifications.
Tools: n8n, OpenAI, Supabase, Gmail API
Explore the portfolio projectsQuestions about AI automation engineering
Can an existing business process be automated without replacing its tools?
Often the first step is connecting the tools already in use. Feasibility depends on their APIs, permissions, data quality and the actions required. Share the current process and systems so we can identify which steps are suitable for automation.
Does every automated workflow need an LLM?
No. Fixed rules, calculations and predictable data transformations are usually better handled by conventional code or workflow nodes. Use an LLM where interpreting or generating language adds a clear benefit.
Can I work with Anurag remotely?
Yes. I am based in New Delhi, India and am open to remote freelance projects and AI engineering roles. Contact me with your workflow, existing tools and expected outcome to discuss the scope.
Tell me what you want to build
Share the problem, your existing tools and the outcome you need. We can discuss the scope, constraints and a practical starting point.
Book a project discussionContact Anurag