Job Description: Senior Software Development Engineer
Role: Senior Software Development Engineer (SDE III) — MERN
Company: Hoora Technologies Pvt Ltd
Location: Nagpur (In-office)
Experience: 7 + Years
Work Week: 5 days
Travel: N/A
About the Company
Hoora is a full-stack vehicle care platform built to simplify how India maintains and manages its cars and bikes. Positioned as India’s #1 Car & Bike Care App, Hoora brings every essential automotive need into one seamless ecosystem. From doorstep car and bike wash to end-to-end service, insurance support, challan management, and a curated marketplace for products and accessories, Hoora functions as a complete 360° vehicle companion. Powered by technology, on-demand logistics, and trained professionals, the platform replaces fragmented automotive experiences with one integrated solution. Hoora’s mission is simple: make vehicle ownership effortless, accessible, and reliable, whether it’s routine cleaning, urgent servicing, compliance needs, or upgrading your ride.
Role Summary:
Hoora is building India's first Vehicle 360 platform — doorstep car and bike care, an accessories marketplace, insurance, FASTag and challan services, all in one app. We're in 85+ cities, we've served over a lakh customers, and we're growing roughly 20% month on month on a ₹36 Cr revenue run rate. We're hiring a Senior Software Development Engineer to own a core system of that platform end to end — architecture, implementation, and production. This is a deeply hands-on role. You'll spend most of your week in the codebase, and the rest making the engineers around you better. If you want to write architecture documents and hand them to other people to build, this isn't the role. If you want to own the hard parts of a real logistics and payments platform in a small team where your decisions actually land, keep reading.
Roles & Responsibilities Depth in the stack — 7+ years of professional engineering, with real depth, not breadth-only exposure
Node.js — production ownership at scale. You understand the event loop well enough to explain a latency cliff, know when a workload needs worker threads or a different runtime entirely, handle streams and backpressure correctly, and have profiled a memory leak in production rather than restarting on a cron. MongoDB — schema design for real access patterns, not normalised relational schemas in BSON. Index strategy and the ability to read and explain aggregation pipeline performance. Practical experience of what breaks at scale: unbounded document growth, hot shard keys, write contention. You know when Mongo is the wrong choice and have said so out loud. React — modern React in a large codebase. Rendering performance and how to diagnose it. Deliberate state management choices with an articulated rationale. Our admin, partner and ops consoles are React, and they're used all day by people whose job depends on them. Express / Node API design — contracts that survive versioning, authn/authz, request validation, idempotency, and rate limiting. TypeScript — across both ends, used to prevent classes of bug rather than to satisfy a linter.
Distributed systems and operational judgment
Designed and operated systems with real reliability requirements: retries and backoff, idempotency, timeouts, circuit breakers, graceful degradation. Async processing at scale — queues, workers, event-driven patterns. Our peak is sharp and daily; the system has to absorb it. Caching strategy, and honesty about invalidation. Observability as a design input: structured logging, metrics, tracing. You've debugged something you couldn't reproduce locally. CI/CD, containers and cloud infrastructure to the depth of someone who owns their own deployments.
Ownership and influence at SDE - 3 scope
You've owned a system spanning multiple services and multiple engineers' work, over multiple quarters. You've led a migration or re-architecture in a live system without a downtime window. You write design documents a strong peer can genuinely critique, and you've changed your mind because of one. You've mentored engineers who visibly grew, and you can name what you did. You've held a technical position under pressure — and you've also lost an argument and committed to the other decision without sabotaging it.
AI-assisted engineering
You use coding agents as a serious part of your workflow, and review generated output with the same rigour you'd apply to a colleague's PR. You can explain and defend every line you ship, regardless of what wrote it. You know where AI assistance degrades quality — novel logic, payment paths, subtle concurrency — and switch modes deliberately.
What you'll own
Rebuild assignment and capacity. Slot capacity by city and pin code, partner matching, ETA, and — the hard part — graceful reassignment when a partner cancels 20 minutes before the slot. This is the single highest-value engineering problem at Hoora. Own the payments path. Prepaid orders, wallet, refunds, partner payouts, and reconciliation against the gateway. Idempotency and auditability are not optional here. A refund that silently fails is a customer we lose permanently and a review we can't undo. Make the expensive decisions. Data modelling, service boundaries, consistency guarantees, build-vs-buy. You'll make calls that are hard to reverse, on systems handling money, and you'll be accountable for them. Own it in production. SLOs, dashboards, alerting, on-call, and the systemic fixes for recurring incident classes. Right now much of this doesn't exist. You'd build it. Scale the multi-city architecture. At 20% MoM growth, the design that works at 85 cities won't work at 200. Getting ahead of that is part of the job. Integrate the Vehicle 360 surface. Insurance, FASTag, challan and vehicle-history providers — third parties with inconsistent SLAs, flaky APIs and no interest in your uptime. Abstraction and failure design matter more than the integrations themselves. Raise the bar through review. Your code review comments will be one of the highest-leverage things you do here. Grow engineers. Mentor and sponsor our SDE Is and IIs. Their promotion cases should be stronger because of your investment. Interview, and raise the hiring bar. Set the AI-native standard. Define how the team works with coding agents — patterns, review expectations, verification discipline, and honest judgment about where they degrade quality. Separately, there's real AI product surface here: demand forecasting for slot capacity, partner allocation, and image-based service quality verification. You'd be close to all of it.
