Skip to content
View venkatchittoor's full-sized avatar

Block or report venkatchittoor

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
venkatchittoor/README.md

Hi, I'm Venkat 👋

Senior Consultant at Slalom | Data Engineering + Agentic AI | Oil & Gas Domain

I build data pipelines and agentic AI systems — from production Medallion architectures to multi-agent crews that reason across real-world domain data. My portfolio spans the full spectrum from scripted pipelines to autonomous tool-calling agents, with a focus on the oil & gas domain where AI meets decades of engineering expertise.

🗂️ Portfolio Index

Repo Category Pattern Novel Aspect
sales-intelligence-pipeline Data Engineering Pipeline Built entirely via Claude Code — no manual code written
ecommerce-pipeline Data Engineering Pipeline 6-phase production stack — streaming, DLT, NL querying, scheduled jobs
iceberg-lakehouse-comparison Data Engineering Comparison Same pipeline built twice — Delta Lake vs Apache Iceberg side-by-side
data-incident-agent Agentic AI Prompt-based Monitoring only — never acts, reports and escalates with reasoning
pricing-decision-agent Agentic AI Prompt-based Confidence-gated autonomy — HIGH acts instantly, MEDIUM/LOW defers to human
customer-behavior-crew Agentic AI Prompt-based crew LLM-driven routing via agent registry — extensible by addition
drilling-npt-agent Agentic AI + O&G Prompt-based 20 years of drilling expertise encoded in prompts — not hardcoded rules
offset-well-intelligence-crew Agentic AI + O&G Prompt-based crew Real Equinor Volve data — cross-parameter formation evaluation reasoning
tool-calling-dq-agent Agentic AI True tool-calling Claude writes its own SQL and drives the investigation loop autonomously
volve-drilling-advisor Agentic AI + O&G Constrained tool-calling Cross-project — offset well formation intelligence feeds live drilling advisory
volve-drilling-advisor-strands Agentic AI + O&G Strands SDK Same advisor rebuilt on AWS Strands — @tool decorator replaces ~200 lines of agent infrastructure

🧭 The Agentic AI Spectrum

Prompt-based          Constrained              True tool-calling       Strands SDK
(prior agents)        tool-calling             (tool-calling-dq-agent) (volve-drilling-advisor-strands)
      |               (volve-drilling-advisor)         |                       |
Your code decides   Claude decides             Claude decides          Claude decides
the sequence        which tool + args          which tool + args       which tool + args
                    Queries pre-written        AND writes its own SQL  Framework handles
                                                                       all infrastructure

🛢️ O&G Domain Expertise

Four projects apply agentic AI to oil & gas — a domain where the reasoning inside the prompts is the moat, not the architecture:

  • drilling-npt-agent — detects stuck pipe, washout, bit balling, twist-off from live sensor streams using cross-parameter reasoning
  • offset-well-intelligence-crew — correlates formation tops, assesses reservoir quality, and forecasts drillability from real Volve field well logs
  • volve-drilling-advisor — delivers real-time drilling advisories by combining live F-15 drilling parameters with formation intelligence from offset well Silver tables
  • volve-drilling-advisor-strands — same advisor rebuilt on AWS Strands SDK — demonstrates how an agentic framework eliminates boilerplate without touching domain logic

🛠️ Tech Stack

Python Databricks Apache Spark Delta Lake Apache Iceberg Claude API AWS Strands

📬 Connect

LinkedIn GitHub

Popular repositories Loading

  1. sales-intelligence-pipeline sales-intelligence-pipeline Public

    End-to-end sales analytics pipeline built with Claude Code and Databricks Data Intelligence Platform

    Python

  2. ecommerce-pipeline ecommerce-pipeline Public

    e-commerce data pipeline with Medallion Architecture, streaming, and anomaly detection on Databricks

    Jupyter Notebook

  3. iceberg-lakehouse-comparison iceberg-lakehouse-comparison Public

    Apache Iceberg vs Delta Lake — same Medallion pipeline built on two different table formats

    Python

  4. data-incident-agent data-incident-agent Public

    Autonomous AI agent that monitors a multi-layer Databricks lakehouse (Bronze/Silver/Gold), reasons about anomalies using Claude API, and files structured incident reports — without hardcoded rules …

    Python

  5. pricing-decision-agent pricing-decision-agent Public

    Autonomous AI agent that analyzes revenue and return signals across product categories, reasons about pricing opportunities, and self-determines when to act — auto-applying HIGH confidence price ad…

    Python

  6. customer-behavior-crew customer-behavior-crew Public

    A multi-agent AI crew for customer behavior analysis — an Orchestrator routes business questions to specialist Data and Business agents, synthesizes their outputs, and returns a single coherent ans…

    Python