Fractional CTO & Independent Backend Engineer — Neeraj Garg, Delhi Code is 30%.
User pain
is the other 70%.

Joined as the first in-house backend engineer and took the entire stack over from a third-party agency. Scaled the consumer site, shipped the mobile site, integrated payments — the engagement that hardwired my zero-to-one instincts. Hiring a backend engineer as your first in-house technical hire is one of the highest-leverage decisions a startup makes; I've been on both sides of that decision.

At Dineout I learned to watch production fail at 2am, understand why, and fix it before morning — not because a manager asked, but because the system was mine. That sense of full-stack ownership, from architecture to on-call, is what I bring to every 0 to 1 product engineering engagement today.

• End-to-end ownership of www.dineout.co.in backend
• Built m.dineout.co.in (mobile site)
• PayU integration · Facebook API · affiliate booking widgets
• REST APIs for third-party integration

Joined NuvoEx at a time when logistics software in India was being built from scratch — no mature playbook, just the operational reality of moving goods at scale. I worked across the full logistics stack: reverse logistics, forward dispatch, and warehouse quality control, each a distinct system with its own data model and failure modes.

Logistics is where you learn that edge cases aren't edge cases — they're the job. A reverse pickup that fails silently costs money and trust. A warehouse QC system that's wrong even 1% of the time compounds into real inventory losses. Building reliable systems in this domain sharpened my instinct for correctness over cleverness, and for designing workflows that operators can actually trust under pressure.

• Reverse logistics system — pickup tracking & exception handling
• QC warehouse management system
• Forward logistics dispatch system
• Led and managed a team of multiple developers

Designed and built a real-time booking platform for car servicing, washes, and roadside breakdowns — Uber-style matching between mechanics and consumers, from a blank repo. This was 0 to 1 product engineering in the strictest sense: no existing codebase, no prior art, just a founding vision and a market that needed it yesterday.

As CTO and the sole backend engineer for the first six months, I owned every architectural decision. The geo-matching system, surge pricing logic, and dual-sided booking flow all had to be right from the start — a marketplace can't easily rewrite its core matching engine mid-flight. The choices I made in week one were still running in production a year later.

• Whole-system architecture for consumer + mechanic dual-app
• Real-time dispatch, surge pricing, geo-matching
• Managed multi-developer team across iOS, Android, backend

Inherited a Ruby on Rails backend that worked but couldn't scale. Re-architected the whole thing for performance, scalability, and maintainability without freezing the product team. Classic Rails backend re-architecture challenge: built to get to market, not to grow past it.

The hardest part of this engagement wasn't the code — it was sequencing the work so the product team never had to pause their roadmap for an engineering project. I identified which parts of the monolith to extract first, designed the new domain boundaries, and ran the migration without a single big-bang cutover. The result was a codebase the team could actually hire into.

• End-to-end RoR backend re-engineering
• Introduced eCommerce + Subscription modules from scratch
• Led front-end and back-end engineering teams

Stepped into an AVP role at Wildnet Technologies, a digital-services firm delivering tech solutions across SEO, performance marketing, and software for a range of clients. The role sat at the intersection of engineering leadership and client delivery — owning technical direction, team performance, and output quality simultaneously.

Moving from individual contributor to AVP taught me that the leverage of a senior leader isn't in writing better code — it's in creating the conditions where the team does. Setting technical standards, unblocking engineers, and aligning delivery timelines with business commitments are harder problems than any system design challenge, and this engagement put them front and centre.

• Engineering leadership across multiple client delivery teams
• Technical direction and delivery ownership at AVP level
• Raised engineering standards and team performance

Long-running embedded engagement as an independent technical partner to a capital-markets fintech. Python + Django + PostgreSQL on AWS — the full backend stack for a platform where correctness isn't optional. The longest single engagement of my career, and the one that taught me the most about compounding architecture decisions.

Five years means you live with your own decisions. Every abstraction I introduced in 2020 has to still make sense today. Capital-markets work sharpens your discipline fast: data integrity failures have real financial consequences, deployment safety isn't negotiable, and "it works on staging" isn't a release criterion.

• Production engagement ongoing since 2020
• Algorithmic execution & data pipeline work
• Compounding architecture decisions, paid off compounding too