Engineering & AI transformation leader

PANKAJ PANDEY

AVP — Engineering, BYLD Group · Gurugram, India

Fifteen years of engineering, pointed at the business problem first. I built an engineering organisation from five people to over a hundred, and the platforms that team ships support operations contributing more than ₹1,000 crore in revenue.

These days the question I keep returning to is not what AI can do, but which manual, expensive, error-prone piece of a real business it can responsibly remove. This site is where I work that out in the open.

See what I have built The record

15+ Years in technology leadership

100+ Engineering org built and scaled, from 5

₹1,000+Cr Revenue supported by those platforms

10+ Years leading engineering at BYLD

HOW THE WORK COMPOUNDS

Five steps, in this order, every time

01

Find the business problem

Before any technology is named. Which step is manual, expensive or error-prone, what it costs today, and what "fixed" is worth.
02

Decide what not to build

Every initiative is weighed against business value, cost of delay, effort, risk and strategic fit before it enters the engineering queue. Most of the value is in what gets cut.
03

Give the team an operating model

Org design, role clarity, delivery process and engineering metrics — so throughput comes from structure rather than from pressure or headcount.
04

Automate the step that deserves it

AI, automation or a plain script — whichever is proportionate. Humans stay on judgement calls and exceptions; the repetitive rule-based work goes.
05

Measure what changed

Cost, cycle time, error rate, headcount pressure. If the number did not move, the initiative was wrong and it gets said out loud.

THE TRANSFORMATION STORY

Built the platforms behind ₹1,000+ crore of business

Not a headcount number — the engineering organisation and the platforms it shipped, under one leadership mandate.

ENGINEERING LEADERSHIP · BYLD GROUP

From five engineers to a 100+ person organisation

Built and scaled the engineering function from the ground up — engineering managers, tech leads, QA leads and product managers — supporting FMCG, staffing, workforce management, HR and learning businesses including Love In Store, YOMA Business Solutions and Global Checks.

The part that is easy to miss: none of it was a greenfield rewrite. It was done while the platforms those businesses run on stayed live.

Team building Architecture DevOps Delivery process Technology strategy

5 → 100+ Engineers led, from founding the function

₹1,000+ Cr Revenue contributed by the businesses these platforms support

Operating philosophy

"Should we build it?" — not "can we build it?"

Don't implement AI because it is AI. Find the business problem first, then use technology, automation or AI where it creates measurable value. It is easy to say and awkward to hold to, because the incentive always runs the other way — so this site holds to it in public, and the next section is where you can check that.

THE PROOF, NOT THE CLAIM

Three decisions you can check

Anyone can write a philosophy on an About page. These are three places on this site where it cost something to follow — each one is two clicks away.

A summariser with no AI in it

The obvious build was a call to an LLM. The problem did not need one — so it is TextRank, graph theory from 2004. It runs in milliseconds, costs nothing per use, and cannot hallucinate, because every sentence it returns came out of your document. An LLM could not have promised that.

See it work →

Tools that admit what they are not

The Prompt Testing Kit says on the page that it never runs your prompt against a model — it produces, organises, compares and scores, and the judgement stays with you. Overselling it would have been easier and would have made it look broken the first time someone expected magic.

See the boundary →

Sample code that sends you elsewhere

Every page in Code Recipes ends with a link to the vendor's own documentation, and prints the date the samples were last checked against it. Our page goes stale; theirs does not. Saying so loses the click and keeps the reader.

Open Code Recipes →

WHERE THE IMPACT SHOWS UP

Cost, automation and productivity

The pattern across engagements: find the manual or expensive step, decide whether technology should touch it at all, then measure what changed.

Technology cost optimisation

Infrastructure, cloud spend, database load and licensing reviewed against what the business actually needs — then the gap between the two removed. Usually the fastest money on the table.

AI-powered process automation

Repetitive, rule-based verification work taken off manual teams with AI-assisted checks — with humans kept squarely on the judgement calls and the exceptions, which is where they were always worth more.

Engineering productivity

Process, roles and prioritisation restructured so a team ships more on the same or leaner headcount — through clarity and automation, not pressure.

DPDP & DATA GOVERNANCE

Privacy built in, not bolted on

As data platforms scale, DPDP readiness has to live in the engineering, not in a policy document written after the fact. Retrofitting it is where the cost is.

I work with teams on the technical side of that: access control design, data lifecycle and retention practice, privacy-aware architecture, and engineering processes that hold up under a compliance review.

As a technology partner, not a legal advisor. The line matters, and stating it is part of the job.

WHAT THAT COVERS

DPDP readiness Privacy by design Access control design Data lifecycle & retention Secure engineering practice Audit-ready process

This site is the smallest possible version of the same idea: no analytics, no tracking, and nothing you type into a tool stored anywhere.

THE RECORD

Fifteen years, six companies

2016 — now

Associate Vice President, Engineering

Global Checks AI Pvt. Ltd. — part of the BYLD Group

Leading engineering and technology transformation across the BYLD ecosystem — platforms for FMCG, staffing, workforce management, HR and learning businesses, including Love In Store, YOMA Business Solutions and Global Checks.

Grew the engineering organisation from 5 to 100+ — engineering managers, tech leads, QA leads and product managers. The platforms built under that team support operations contributing over ₹1,000 crore in revenue. Current focus is AI implementation and intelligent automation against measurable operational problems.
2017 — 2026

Head of Engineering shared role

Love In Store

A shared engineering leadership responsibility held on BYLD payroll rather than a separate employment — worth stating plainly, because an overlapping date range on a CV usually is not.

Owned end-to-end delivery for Love In Store's FMCG and FinTech portfolio: architecture, product engineering, QA, releases and technology strategy, plus the engineering practices around CI/CD, scalability and cost that the shared function ran on.
2014 — 2016

Software Engineer

VAG Technologies Pvt. Ltd.

Led a team of two to five developers building custom e-commerce on NopCommerce — cart, catalogue and checkout in ASP.NET MVC, payment gateway and ERP integrations, multi-vendor and custom pricing rules.
2013 — 2014

Software Engineer

SunTec India

Client-specific NopCommerce plugins and modules, payment gateway and ERP integration, and the pricing and promotion logic behind several storefronts.
2013

Software Engineer

Inveera Technologies

Custom ERP modules for US real-estate clients — property and lease management, accounting and billing, broker workflows — with REST APIs keeping the ERP in step with CRM and document systems. Requirements gathered directly from US stakeholders.
2011 — 2013

Associate Software Engineer

Acube Promotion House Pvt. Ltd.

A unified messaging platform across SMS, email and two-way shortcode, integrating telecom carrier APIs. Built the queue-backed pipelines that let it dispatch at high volume and low latency — the first system where throughput, not features, was the hard part.

WHERE MY ATTENTION IS NOW

AI, where it earns its place

The question I keep returning to is not what AI can do — it is which manual, expensive, error-prone piece of a real business it can responsibly remove.

AI-powered business solutions Intelligent automation Agentic workflow design AI ROI assessment AI-assisted software engineering Workforce & staffing technology FMCG field-force technology HR & workforce platforms Product engineering & platform architecture Engineering team building & scaling DevOps & engineering productivity Data-driven business platforms

EDUCATION

Where the AI half came from

The two recent programmes are deliberate. Leading AI work without having done the maths is how an organisation ends up buying demos, and the fastest way to stop doing that is to go and learn the subject properly.

Institute of National Importance

Indian Institute of Technology, Mandi

PG Certification — Generative AI & Multi-Agent Systems

2025 — 2026 In progress

Institute of National Importance

Indian Institute of Management, Kozhikode

Professional Certificate Programme in Data Science and Artificial Intelligence for Managers

Sep 2024 — Jun 2025 Completed

eMDP (Asynchronous) · Programme completion certificate

The degree underneath

APJ Abdul Kalam Technological University

B.Tech, Information Technology

2006 — 2010

OFTEN ASKED ABOUT

Data migration OCR Third-party integration

CERTIFICATIONS & WORKSHOPS

Certified, not just attended

A DiSC Certified Practitioner credential, Blanchard's SLII, and Crucial Learning's Crucial Conversations — alongside BYLD Group's Extra Mile Award.

AI & Data
AI Tools & Claude Workshop

be10x

Sep 2026
AI & Data
IIT Mandi iHub — AI Agents and Multi-Agent Systems

Coding Ninjas

Aug 2026
AI & Data
Introduction to Prompt Engineering for Generative AI

LinkedIn Learning

Aug 2026
Engineering
Microsoft Azure Essentials

LinkedIn Learning

Aug 2026
AI & Data
IIT Mandi iHub — Basics of NLP and RAG

Coding Ninjas

Jun 2026
AI & Data
IIT Mandi iHub — Prompt Engineering and API Integration

Coding Ninjas

Jan 2026
Leadership
SLII® — Situational Leadership II

Blanchard

Sep 2025
Delivery
IT Project Management: Delivering Successful IT Projects

Udemy

Jul 2025
Leadership
The Power of Habit

Crucial Learning

Jul 2025
Leadership
Art of Communication

BYLD Group

Jan 2025
Engineering
Create a Python Powered Chatbot in Under 60 Minutes

Udemy

Aug 2024
AI & Data
Building Interactive Apps with Shiny and R

Udemy

Jun 2024
AI & Data
R Programming — R Language for Absolute Beginners

Udemy

Jun 2024
Delivery
Manage Your IT Project with Key Templates and Checklists

Udemy

May 2024
Leadership
Negotiation Skills Workshop

BYLD Group

Mar 2024
Leadership
Getting Things Done

Crucial Learning

Feb 2024
Leadership
Interview Skills (BEI) Workshop

BYLD Group

Jan 2024
Leadership
Critical Thinking Workshop

BYLD Group

Sep 2023
Engineering
Certified DevOps Associate

Cloud Train

Feb 2023
Leadership
Crucial Conversations for Mastering Dialogue

Crucial Learning

Dec 2022
Leadership
DiSC Certified Practitioner

BYLD Group

Nov 2022
Engineering
Google Cloud Foundations

Google Cloud
Recognition
The Extra Mile Award

BYLD Group
Delivery
WhatsApp Marketing & Automation Workshop

BYLD Group