NMNicklesh Madoori Get in touch →

Engineering leader · AI transformation · Author

Turning complex engineering into measurable business value.

I build and scale engineering organizations and enterprise platforms where reliability, business outcomes, and execution have to hold up at scale. Today I lead Subscriptions engineering at Walmart Global Tech and build applied-AI systems on the side.

Outcomes at scale

$60M+

Annual GMV from Subscriptions initiatives

Walmart Global Tech
3×

Increase in subscriber acquisition

Walmart Global Tech
40M+tx/day

Transactions processed daily by systems my teams built

eBay AML · FCC
25M+/day

Daily risk evaluations

eBay AML · FCC
$15M+

Cost avoidance delivered

eBay AML · FCC
1,200→<100

Production defects in 10 months

LVIS

Who I work with

Useful where scale, complexity and outcomes meet.

Three groups get the most from my work. Each has its own way in.

For hiring teams

Companies scaling engineering

Director, VP and CTO-level scope: leading through leaders, platform ownership, and execution discipline across large organizations.

See the impact →
For engineering leaders

Teams adopting AI in delivery

Moving from isolated prompts to repeatable, governed AI workflows with guardrails, specs and measurable value.

See the AI work →
For founders

Builders shipping products

Architecture, practical AI builds, rapid prototyping and the product decisions behind seven platforms I've shipped.

See the builds →

How I work

Lead. Transform. Build.

Strong engineering leadership isn't about personally solving every problem. It's about building the people, systems, standards and feedback loops that let an organization solve the right problems consistently.

Lead.

Scaling teams and building leaders

  • Orgs of 12–25 engineers, end to end
  • 25 engineers across three continents
  • Governance across 300+ engineers
  • Engineers grown into people managers
Transform.

AI-enabled engineering execution

  • Jira, Confluence and GitHub signals turned into coaching
  • Spec-driven code generation and testing
  • Reusable AI skills as a team norm
  • Agents with approvals and controls
Build.

Platforms and products that hold up

  • Systems at 40M+ transactions/day
  • Five deployment pipelines unified into one
  • Seven independent AI platforms
  • 25+ AI agents built
  • 10+ production workflows in n8n

Selected impact

Three problems, three outcomes.

Each case reads as context, the decisions that mattered, and the measured result.

Walmart Global Tech · 2024–now

Growing recurring commerce for Walmart+

ContextFull-stack Subscriptions org covering recurring orders, InHome delivery, Savings & Benefits and Auto Replenishment.

DecisionsEnd-to-end ownership of architecture, roadmap and quality, plus horizontal owners for operational excellence, business metrics and app architecture.

$60M+ · 3×

Annual GMV contribution and growth in subscriber acquisition.

eBay · 2017–2023

Engineering for anti-money-laundering at marketplace scale

ContextFour cross-functional teams, 25 engineers on three continents, supporting AML and Financial Crimes Compliance.

DecisionsPlatform modernization, security and framework upgrades, and continuity across the broader risk organization.

40M+ tx/day

25M+ daily risk evaluations and $15M+ in cost avoidance.

LVIS · 2023–2024

A quality turnaround for a regulated medical product

ContextEngineering, Product, DevOps, ML Ops, QA, UX and Operations across the US and South Korea.

DecisionsClear ownership, quality controls, visible metrics and release discipline. Five deployment pipelines consolidated into one.

1,200 → <100

Production defects within 10 months, while advancing regulatory milestones.

AI transformation

From prompts to governed engineering practice.

Agents alone don't make an operating model. The work is turning experiments into repeatable practice with controls and measured value.

  • 01

    Engineering execution platformReal-time Jira, Confluence and GitHub signals for visibility and data-informed coaching across a ~40-engineer organization.

  • 02

    Specification-driven developmentRepository-based specs that govern generated code, LLM output, tests and coverage standards.

  • 03

    Reusable skills by defaultAny task done twice becomes a reusable AI skill the team can share.

  • 04

    Agents for security and qualitySonar remediation, test coverage, upgrades and SAST/DAST checks, with human approvals built in.

MODEL · 01

Agentic Engineering Operating Model

How people, workflows, agents, tools, policies and feedback loops work together in software delivery.

Working model
MODEL · 02

Agent Control Plane

Permissions, execution, observability, policy, approval and recovery for agents.

Working model
MODEL · 03

AI Engineering Maturity Model

From prompts, to reusable skills, to loops, to agent graphs, to an organizational operating model.

Working model
MODEL · 04

Engineering Effectiveness Model

Quality, delivery, operational health, customer outcomes and AI-enabled change in one view.

Working model

Credentials: GenAI for DevOps Practitioners · GitLab Duo AI Pair Programming · Generative AI for Executives · Responsible AI · Introduction to LLMs.

Independent builds · since 2025

Seven platforms, built hands-on.

Applied-AI R&D into where automation creates real value, where model behavior gets unreliable, and what controls keep AI systems safe.

Also built: a market-analysis engine that scores equities and crypto on 12+ signals, and a daily AI and market-intelligence briefing. In total: 25+ AI agents and 10+ production workflows running in n8n.

Agentic workflowsMulti-LLM orchestrationScheduled agentsReal-time data ingestionAPI integrationn8nFull-stack

Blog · weekly

AI agent adoption, one note a week.

Short, sourced notes on how organizations are adopting AI agents and what it means for engineering leaders.

Writing

Field notes from building and leading.

Practical, opinionated and grounded in real systems. Published on LinkedIn.

Featured series · 18 parts

Specification-Driven Development

How clear specs make engineering, and AI-assisted delivery, dependable.

  • — Useful vs. vague requirements
  • — Non-functional requirements
  • — Architecture and testing
  • — AI-assisted delivery
  • AI in Practice

    Agent workflows, implementation patterns, evaluation and governance.

    Series
  • Field Notes

    Lessons from leading engineering organizations.

    Series
  • Architecture Decisions

    Systems, trade-offs, scalability, reliability and security.

    Series
  • Leadership Frameworks

    Managing up, across and down; IC to manager; manager to director.

    Series
  • Build Logs

    Decisions and lessons from independent product work.

    Series

About

A career in systems that can't fail quietly.

From marketplace risk and compliance, to regulated MedTech, to recurring commerce at Walmart scale. Alongside it, hands-on AI building and writing.

IndependentWalmart Global TechLVISeBayToday
Education · B.E., Mechanical Engineering, Osmania UniversityTop skills · Transaction monitoring · Financial crimes compliance · Solution architecture

Advisory

Focused engagements for engineering leaders.

Limited availability each month. Each engagement has a defined scope and a concrete deliverable.

Assessment

AI Engineering Effectiveness

Current-state review, prioritized gaps, metrics and governance recommendations, and a 30/60/90-day roadmap.

Assessment

Agentic Workflow Review

Workflow map, automation opportunities, a risk and control review, and a prototype plan with success criteria.

Advisory

Engineering Operating Model

Ownership, quality gates, operational excellence, reporting and leadership mechanisms.

Development

Engineering Leader Development

A structured growth plan for engineers and managers stepping into bigger leadership roles, including turning your track record into clear positioning, proof points and a portfolio site.

Contact

Let's solve the hard part together.

Emailnicklesh@gmail.com
LocationSan Francisco Bay Area