---
title: Bhanuprakash Avadutha — AI Agent & Automation Engineer
site: bhanu.ai
canonical: https://bhanuavadutha.in
location: Hyderabad, India
updated: September 2026
---

# Bhanuprakash Avadutha — AI Agent & Automation Engineer

I build AI agents and automation systems that turn complex financial workflows into reliable infrastructure.

An AI agent and automation engineer with deep Indian financial-market context, building production systems across data, research, compliance, reporting, and operations.

## Verified impact

- **84% Editorial agreement:** AI financial-news filtering measured against human editorial decisions
- **8 Production systems:** Shipped across data, research, compliance, onboarding, and reporting
- **~1,500 Hours saved annually:** Estimated recurring manual effort removed across operational workflows

## What I build

- **Agent systems:** Tool-using workflows, grounded reasoning, structured outputs, human review, and evaluation loops.
- **Automation engineering:** Reliable browser, document, email, and operational automations designed around failure and recovery.
- **Data platforms:** Cloud ingestion, transformation, quality contracts, orchestration, and analytics-ready data models.
- **Financial workflows:** Applied knowledge of Indian markets, research operations, KYC, compliance, reporting, and broker formats.

## Selected systems

### Financial News Intelligence

A production pipeline that turns a noisy overnight market-news stream into a focused, reviewable editorial queue.

**Outcome:** An orchestrated ingestion, classification, review, and publishing flow with explicit evidence and human oversight.

**Impact**
- 4–5 hours → 8 minutes
- 2 overnight operators → 0
- 84% editorial agreement

**Stack:** Python, Airflow, Gemini, GCP, BigQuery, Terraform

[Case study](https://bhanuavadutha.in/work/financial-news-intelligence) · [GitHub](https://github.com/BhanuprakashAvadutha/indian-financial-news-pipeline)

### Research Report Factory

A controlled reporting system that turns structured stock research into validated, chart-rich client documents overnight.

**Outcome:** Reusable templates, chart generation, validation gates, and overnight batch production with review-ready outputs.

**Impact**
- 120–200 DOCX reports
- 5–7 days → overnight
- Automated validation gates

**Stack:** Python, DOCX, Pandas, Matplotlib, Validation

[Case study](https://bhanuavadutha.in/work/research-report-factory) · [GitHub](https://github.com/BhanuprakashAvadutha/stock-research-report-automation)

### Broker Order Orchestration

A structured relay that converts investment instructions into broker-ready order artifacts with validation and auditability.

**Outcome:** Schema-validated transformation from natural language to a 21-column broker format in seconds.

**Impact**
- Instruction → CSV in ~10 sec
- 21-column schema
- Designed for multi-client operations

**Stack:** Next.js, Claude, Zod, TypeScript, CSV

[Case study](https://bhanuavadutha.in/work/broker-order-orchestration) · [GitHub](https://github.com/BhanuprakashAvadutha/basket-csv-generator)

### GCP Market Data Platform

A medallion-style financial data platform for reproducible ingestion, transformation, quality checks, and analytics.

**Outcome:** Bronze, Silver, and Gold layers with infrastructure as code and analytics-ready outputs.

**Impact**
- Bronze / Silver / Gold
- Infrastructure as code
- Analytics-ready models

**Stack:** GCP, BigQuery, dbt, Terraform, Power BI

[Case study](https://bhanuavadutha.in/work/gcp-market-data-platform) · [GitHub](https://github.com/BhanuprakashAvadutha/gcp-data-platform)

### Client Onboarding Pipeline

A compliance-aware onboarding flow connecting forms, documents, email, CRM updates, and operational handoffs.

**Outcome:** A single automated workflow with document creation, notifications, and traceable compliance steps.

**Impact**
- 2–3 days → under 5 min
- Automated handoffs
- Compliance-aware records

**Stack:** Apps Script, Google Drive, Gmail, CRM, Automation

[Case study](https://bhanuavadutha.in/work/client-onboarding-pipeline) · [GitHub](https://github.com/BhanuprakashAvadutha/client-onboarding-pipeline)

### Trendlyne Research Automation

A session-isolated toolkit for collecting and validating structured research data across large stock lists.

**Outcome:** Typed extraction, session isolation, rate control, and cloud-ready storage for repeatable research operations.

**Impact**
- 50–200 stocks per run
- Session isolation
- Typed validation

**Stack:** Python, Playwright, Pydantic, GCS, BigQuery

[Case study](https://bhanuavadutha.in/work/trendlyne-research-automation) · [GitHub](https://github.com/BhanuprakashAvadutha/trendlyne-scraper-toolkit)

## Experience

### Data Engineer · AI Automation Builder — 2024 — Present

SEBI-registered financial research firm. Sole engineer responsible for production data and automation systems spanning editorial intelligence, research, client operations, compliance, and reporting.

- Shipped eight production systems
- Designed cloud data and document pipelines
- Embedded AI where rules alone were insufficient

### Master of Computer Applications — 2024

Postgraduate degree. Formal computer-applications foundation strengthened through production engineering and financial-domain work.

- Software engineering
- Data systems
- Applied AI

## Skills

- Python
- Machine Learning
- LangChain
- LangGraph
- LLM APIs
- RAG
- Agent workflows
- TypeScript
- React
- Next.js
- Tailwind CSS
- SQL
- Pandas
- Airflow
- BigQuery
- dbt
- GCP
- Supabase
- Vercel
- Terraform
- Git
- GitHub
- Playwright
- Pydantic
- Zod
- Apps Script

## Availability

Open to consulting, partnerships, and high-impact engineering roles.

## Links

- Portfolio: https://bhanuavadutha.in
- Work: https://bhanuavadutha.in/work
- Contact: https://bhanuavadutha.in/contact
- GitHub: https://github.com/BhanuprakashAvadutha

This document contains only approved public portfolio information. No testimonials or client identities are inferred.