Hire MERN Stack Developer India 2026: ₹40K-₹3L/mo Real Pricing
Hire MERN Stack Developer India 2026: The Honest Pricing Guide
If you are searching "hire MERN stack developer India," you are probably balancing three concerns: real cost, talent quality, and how to avoid getting burned by the cheap freelancer who ghosts mid-project. We have seen all three play out across 200+ MERN engagements at Codingclave since 2017.
I am Ashish Sharma, founder of Codingclave — a Top Rated Upwork agency from Lucknow, India. We have placed 70+ MERN stack engineers with US, UK, Australian, Gulf, and Indian clients in the last 4 years. This guide is the unfiltered version: real 2026 INR rates, honest comparison vs other regions, the hiring playbook that actually works, the interview questions that filter quickly, and the red flags that scream "this hire will fail."
TL;DR — Real 2026 Indian MERN Stack Developer Pricing
| Experience Tier | Monthly Rate (Direct) | Through Agency | What You Get |
|---|---|---|---|
| Junior (1-2 years) | ₹40K-₹65K | ₹60K-₹95K | Code under guidance, basic React + CRUD APIs, learning |
| Mid (3-5 years) | ₹80K-₹1.5L | ₹1.1L-₹1.9L | Independent execution, owns features end-to-end |
| Senior (5-8 years) | ₹1.5L-₹2.4L | ₹1.9L-₹3L | System design, mentors juniors, owns architecture |
| Lead (8+ years) | ₹2.4L-₹3.5L | ₹2.8L-₹4.5L | Tech leadership, hiring, multi-team coordination |
| Top 1% Senior | ₹3L-₹5L | ₹4L-₹6L | Ex-Big Tech, ex-funded startup, pure A-player |
For perspective: a comparable mid-senior MERN developer costs ₹2L-₹4L/mo in Eastern Europe, ₹1.8L-₹3.5L/mo in LATAM, ₹4L-₹8L/mo in Singapore/Dubai, and ₹6L-₹12L/mo in the US. India remains the single most cost-effective skilled MERN market in 2026.
What Makes Indian MERN Stack Developers Strong (and Where the Gaps Are)
Strengths
Largest English-speaking JavaScript developer pool globally — industry estimates put the MERN-capable engineer count in India at 5M+, larger than the US, UK, and Western Europe combined. More candidates per role, faster hiring funnel, deeper specialization on every framework.
Mature outsourcing ecosystem — India has been a global outsourcing hub since the late 1990s. Top engineers know how to work with US/UK/AU/Gulf clients across time zones, in a mix of async and sync, with documentation discipline that matches Western expectations.
Gulf time-zone alignment — India is only 1.5 hours ahead of UAE/Saudi/Qatar/Bahrain/Oman. A full 8-hour business-day overlap with Gulf clients makes Indian MERN engineers essentially indistinguishable from local hires on day-to-day communication, at a fraction of the cost.
UK + Europe partial overlap — IST is 5.5 hours ahead of GMT in winter, 4.5 in summer. India 1:30 PM to 6:30 PM overlaps with UK 8 AM to 1 PM — half a working day of real-time overlap is standard.
Cost-efficiency that compounds — at 60-75% cheaper than US equivalents, the savings compound when you hire 3-5 engineers vs 1. A 4-engineer Indian MERN team costs roughly the same as 1 senior US engineer.
Strong English fluency in Tier-1 cities — Bengaluru, Mumbai, Delhi NCR, Pune, Hyderabad, Chennai all have native-fluent professional English. Communication is rarely the bottleneck for clients hiring from these cities.
Honest Gaps
Time-zone friction with US Pacific — the 12-13 hour gap means real-time daily standups need creative scheduling. India 8-9 PM standup overlaps with US 7-8 AM PT — doable, but the engineer has to commit to evening hours.
Variable quality — India has both world-class engineers AND a huge pool of unvetted freelancers selling themselves as senior. Vetting is critical, and skipping it is the #1 reason offshore hires fail.
Tier-2/3 city infrastructure — power outages and ISP reliability issues outside Bengaluru/Mumbai/Hyderabad/Pune still happen occasionally. Most serious agencies (us included) provide backup power + dual ISP for Tier-2 city teams.
Cultural directness — Indian engineering culture is sometimes more deferential than US/EU norms. Some clients prefer engineers who openly push back on bad product decisions; you need to explicitly invite that behavior during onboarding.
The Five Hiring Paths (Ranked by Failure Rate)
Path 1: Vetted Indian Agencies (Codingclave, Tatvasoft, ValueCoders, Mantra Labs, etc.)
Failure rate: under 10% (project does not deliver as promised)
Cost: ₹85K-₹3L/mo per developer — a 30-50% premium over direct freelance that buys you vetting, project management, and bench depth.
Best for: Most clients hiring 1-5 developers, business-critical projects, projects running 3+ months.
Pros: Pre-vetted talent (we test every dev before placing), GST-compliant export invoicing, bench depth (sick or quitting developer = swap, not stoppage), fixed monthly billing is predictable, single contract for multiple developers.
Cons: Higher monthly rate than direct freelance, less direct relationship with the developer because the agency intermediates.
Path 2: Top Rated Upwork Freelancers
Failure rate: around 20% (Upwork's Top Rated and Top Rated Plus tiers filter out the worst)
Cost: ₹80K-₹2.4L/mo equivalent at typical hourly rates.
Best for: Clients with strong in-house engineering management, projects under 3 months, budget-conscious clients comfortable doing their own vetting.
Pros: Direct relationship with the engineer, Upwork escrow + dispute resolution, public ratings + reviews provide social proof, lower management overhead than a direct LinkedIn hire.
Cons: You manage project + performance personally, freelancer getting sick or busy = your project stalls, hourly billing can creep over budget without strict scope control.
Path 3: Toptal / Arc / Lemon.io / Turing
Failure rate: under 5% (deepest vetting in the market, top 3% of applicants)
Cost: ₹2L-₹4L/mo — premium pricing.
Best for: Funded startups, US/UK clients with quality priority over cost, projects where a single delivery miss is catastrophic.
Pros: Strongest vetting on the market, replacement guarantees, dedicated account manager.
Cons: 2-3x premium over Indian agencies for similar engineer quality, US-centric pricing models.
Path 4: Direct LinkedIn / Naukri / Referral Outreach
Failure rate: around 30% (you do all vetting, no platform safety net)
Cost: ₹40K-₹2L/mo (negotiated direct)
Best for: Clients with a deep Indian network, time to vet 10-20 candidates, contracts where Upwork or agency fees would meaningfully hurt margins.
Pros: Cheapest path on paper, direct relationship.
Cons: Time-intensive vetting (10-25 hours minimum), no platform protection, no dispute resolution, contracts must be drafted manually.
Path 5: Full-Time Employment (Naukri, Cutshort, Hirect)
Failure rate: around 25% (depends entirely on your hiring process maturity)
Cost: ₹6L-₹30L/year all-in (salary + 30% benefits + recruiter fees of 8.33%-12.5% of CTC)
Best for: Companies hiring 5+ MERN engineers, building a permanent India team, willing to set up an Indian entity or use an Employer of Record.
Pros: Lowest per-developer rate over a multi-year horizon, dedicated team, IP ownership terms are clearer.
Cons: Need an Indian Private Limited entity or PEO/EOR arrangement, 30-90 day hiring cycle, you become the employer with the full HR + compliance burden, hard to scale down.
What to Look For in a 2026 Indian MERN Stack Developer
The MERN skill bar in 2026 is meaningfully higher than 2022. Here is what separates strong from mediocre candidates.
Must-Have (Mid-Senior Level)
- TypeScript fluency — vanilla JavaScript is a red flag in 2026. TS catches roughly 60-70% of bugs at compile time and is the baseline for any production codebase we accept.
- React 18/19 deep knowledge — hooks, context, Suspense, error boundaries, Server Components (if Next.js), memoization patterns, render performance debugging.
- Node.js + one framework deeply — Express (most common), Fastify (high-perf APIs), NestJS (enterprise structure), or Next.js API routes for full-stack.
- MongoDB at scale — query optimization, aggregation pipelines, indexing strategy, transactions, replica sets, sharding awareness.
- Async patterns — promise composition, error propagation, AbortController, streaming responses.
- Auth flows — JWT + refresh tokens, OAuth2, session management, Clerk or NextAuth integration.
- Docker basics — write a multi-stage Dockerfile, debug container networking.
- Git workflow — feature branches, PR reviews, conventional commits, rebase vs merge tradeoffs.
Strong Differentiators (2026)
- AI integrations — OpenAI or Claude SDK with streaming, function calling, structured outputs.
- Serverless experience — Vercel Functions, AWS Lambda, Cloudflare Workers, edge runtime constraints.
- Modern UI — Tailwind CSS, shadcn/ui, Radix, framer-motion, responsive + accessible component libraries.
- State management — TanStack Query, Zustand, Jotai (Redux is increasingly out of fashion for new builds).
- Queue systems — BullMQ, AWS SQS, Inngest.
- Observability — Sentry, PostHog, OpenTelemetry, Datadog basics.
- Database tooling — Mongoose schemas, Prisma with MongoDB, migration discipline.
- Real-time — Server-Sent Events, WebSockets, Socket.IO, Pusher.
Senior-Level Add-Ons
- System design at scale (1M+ requests/day patterns)
- Performance profiling and optimization on the React side (React DevTools profiling, bundle analysis)
- Multi-region MongoDB deployment + data consistency
- Security: OWASP top 10, dependency scanning, secrets management
- Mentoring + code review skills
Red Flag Phrases on Resumes (Avoid)
- "Expert in 15+ frameworks" — shallow generalist
- "Full-stack developer" with no specific framework depth listed
- "Worked on enterprise systems" with no specifics about scale, team size, or personal role
- Years of experience listed inconsistently across roles
- No GitHub link, or empty GitHub, or only forks and tutorials
12 Interview Questions That Actually Filter MERN Candidates
Skip generic LeetCode puzzles — they do not predict job performance for application engineers. Use these instead.
React (3 questions)
-
"Walk me through how you'd debug a React component that re-renders too often." Strong answer: React DevTools Profiler,
React.memo,useMemo/useCallback, dependency-array hygiene, context-split patterns. Weak answer: vague references to "performance optimization." -
"How do you handle loading, error, and empty states in a list view fetched from MongoDB?" Strong answer: explicit state machine or TanStack Query state flags, skeleton UI, accessible error messages, empty-state CTAs. Weak answer: only the happy path.
-
"Explain Server Components vs Client Components in Next.js App Router." Strong answer: data fetching on the server, no JS shipped for RSC,
use clientboundary, when to split. Weak answer: "Server Components run on the server."
Node.js + Express (3 questions)
-
"Build a rate-limited API endpoint that protects against burst traffic — what does your middleware stack look like?" Strong answer: token bucket or sliding window, Redis-backed counter, per-IP and per-user keys, 429 + Retry-After headers. Weak answer: a single
setTimeoutor in-memory counter. -
"How do you handle errors that originate inside async middleware?" Strong answer:
express-async-errors, centralized error handler, distinguishing operational vs programmer errors, Sentry/observability hook. Weak answer: try/catch everywhere with no central handler. -
"How would you design a webhook receiver for Stripe or Razorpay?" Strong answer: signature verification, idempotency keys, queue + worker pattern, retry strategy, dead-letter queue. Weak answer: synchronous processing inside the request handler.
MongoDB (3 questions)
-
"Walk me through optimizing a slow aggregation pipeline." Strong answer:
explain()output, compound indexes,$matchearly, projection to reduce shape,$lookupcost awareness. Weak answer: "I'd add an index." -
"How do you model a many-to-many relationship in MongoDB?" Strong answer: embedding vs referencing tradeoffs based on read/write ratio, bucket pattern, denormalization for query speed. Weak answer: defaults to SQL-style join tables.
-
"How do you handle multi-document transactions in MongoDB?" Strong answer: replica set requirement,
startSession, retry on transient errors, transaction size limits. Weak answer: assumes Mongo cannot do transactions.
System Design + Real World (3 questions)
-
"Design a multi-tenant SaaS database schema in MongoDB — one DB per tenant vs shared with tenantId field. What do you pick and why?" Strong answer: discusses isolation, backup granularity, query simplicity, cost tradeoffs.
-
"You ship a feature on Friday. By Monday, the engineer who wrote it is on leave and the feature is broken in production. Walk me through your response." Strong answer: revert strategy, observability/logging quality required, on-call runbook, post-mortem discipline.
-
"How do you keep
package.jsondependencies under control across a 12-month project?" Strong answer: lockfile discipline, regularnpm audit+ Dependabot/Renovate, major-version upgrade cadence, removing unused deps. Weak answer: never touched after initial install.
The Vetting Process We Use (and You Should Too)
Our 4-step vetting eliminates 90%+ of unqualified MERN applicants:
Step 1: GitHub + Portfolio Review (15 minutes per candidate)
- Real original projects, not tutorial clones
- Recent commit activity (last 6 months)
- Code quality on display (TypeScript, tests, README discipline)
- At least one project with a live URL you can click
Step 2: Written Async Communication Test (24-hour async)
- Send a small architecture question via email or Slack
- Evaluate: does the response demonstrate understanding, ask clarifying questions, structure thinking?
- This single step filters out 40-60% of candidates
Step 3: Live Coding Session (60-90 minutes)
- Pair-program a real MERN task — for example, "build a paginated React table that fetches from a Node/Express endpoint with MongoDB aggregation"
- Evaluate: code quality, problem decomposition, debugging approach, communication during coding
- Avoid LeetCode-style puzzles — they do not predict job performance for application engineers
Step 4: Paid Trial Week (₹15K-₹25K, 1 week)
- Real ticket from your backlog
- Evaluate: actual delivery, communication cadence, code review responsiveness
- This is the highest-signal step. Many clients skip it because of cost — that is exactly why their offshore hiring fails.
For your own hiring we recommend at minimum: GitHub review + written async test + live coding. Skip any step and your failure rate climbs sharply.
Real Cost Comparison: India vs US vs UK vs Singapore vs Dubai
| Region | Junior | Mid | Senior | Lead |
|---|---|---|---|---|
| India | ₹40K-₹65K (USD 480-780) | ₹80K-₹1.5L (USD 960-1,800) | ₹1.5L-₹2.4L (USD 1,800-2,880) | ₹2.4L-₹3.5L (USD 2,880-4,200) |
| US (Tier-1 cities) | ₹3L-₹5L (USD 3,600-6,000) | ₹6L-₹10L (USD 7,200-12,000) | ₹8L-₹14L (USD 9,600-16,800) | ₹12L-₹22L (USD 14,400-26,400) |
| UK (London) | ₹2L-₹3.5L (USD 2,400-4,200) | ₹4.5L-₹7L (USD 5,400-8,400) | ₹6L-₹9L (USD 7,200-10,800) | ₹9L-₹14L (USD 10,800-16,800) |
| Singapore | ₹2.5L-₹4L (USD 3,000-4,800) | ₹4L-₹6L (USD 4,800-7,200) | ₹6L-₹8L (USD 7,200-9,600) | ₹8L-₹13L (USD 9,600-15,600) |
| Dubai/UAE | ₹2L-₹3.5L (USD 2,400-4,200) | ₹4L-₹6L (USD 4,800-7,200) | ₹6L-₹8.5L (USD 7,200-10,200) | ₹8.5L-₹13L (USD 10,200-15,600) |
India wins on cost across every tier. For mid-senior tier (where most hires happen), India is roughly 60-75% cheaper than US, 60-70% cheaper than UK, and 55-65% cheaper than Singapore/Dubai.
What Indian MERN Stack Developers Build for Foreign Clients
Top use cases we see weekly across active engagements:
B2B SaaS (Most Common)
- Multi-tenant architecture with MongoDB tenant isolation
- REST + GraphQL APIs with Express or Apollo Server
- Auth + RBAC (see our Next.js RBAC guide)
- Stripe / Razorpay subscription flows
- Webhooks + event-driven workflows
- Admin dashboards with React + Tailwind + shadcn/ui
Fintech / Payment Systems
- Payment gateway integrations (Razorpay, Stripe, PayPal)
- KYC + risk scoring with MongoDB document storage
- Reconciliation systems
- Audit logging
- Compliance reporting (PCI DSS, RBI)
AI-Powered Products
- OpenAI / Claude integration with streaming responses
- RAG over customer data with MongoDB Atlas Vector Search
- Multi-step agent workflows
- Voice agents with ElevenLabs
- Embeddings pipelines
E-commerce Platforms
- Shopify / WooCommerce custom storefronts
- Order management systems with MongoDB
- Inventory sync across channels
- WhatsApp commerce (WATI)
Internal Tools + Admin Panels
- Custom admin dashboards with React + Material UI or shadcn/ui
- ETL pipelines from external APIs into MongoDB
- Reporting + analytics dashboards
- Workflow automation tools
Sample Engagement Models We Offer
| Model | Best For | Pricing | Commitment |
|---|---|---|---|
| Dedicated MERN Developer (Monthly) | 3+ months ongoing work | ₹1.1L-₹3L/mo | Min 3 months |
| Project-Based (Fixed Scope) | Clear scope + deliverable | ₹2L-₹25L per project | One-time |
| Hourly (Upwork or direct) | Short tasks, ad-hoc work | ₹1,500-₹3,500/hour | No minimum |
| MERN Squad (2-5 devs + PM) | Complex projects, full team | ₹4L-₹15L/mo | Min 4 months |
| Embedded Engineer | Augment your existing team | ₹1.5L-₹2.8L/mo | Min 3 months |
What's New in Indian MERN Hiring in 2026
1. AI-Assisted Productivity Doubled Per-Developer Output
Senior Indian MERN developers using GitHub Copilot + Cursor + Claude Code ship roughly 1.5-2x faster in 2026 than 2024. We are seeing projects that took 12 weeks in 2023 now ship in 6-8 weeks. Hourly rates did not drop proportionally — it is pure productivity gain for clients.
2. TypeScript Became Mandatory
We no longer place vanilla JavaScript developers for production work. TypeScript proficiency is the new baseline. Candidates with only JS experience need to demonstrate they are actively upgrading.
3. Next.js Full-Stack Eating Classic MERN for SaaS
Most new SaaS hires in 2026 ask for "Next.js full-stack engineer" rather than "MERN developer." Express + separate React app is still common in existing codebases — but for greenfield SaaS, the Next.js App Router pattern collapses the stack and reduces team size.
4. India's Tier-2/3 Cities Producing Strong Talent
Lucknow, Indore, Coimbatore, Kochi, Bhubaneswar, Jaipur, and Chandigarh are producing strong MERN engineers in 2026 at 30-40% lower rates than Bengaluru/Mumbai. Better fiber internet + remote-first culture means Tier-2 talent is now competitive on quality.
5. Time-Zone Coverage Models
Many US clients now hire 2 Indian engineers in different shifts: morning shift (covers US evening), evening shift (covers US morning). Cheaper than one US engineer and provides 16-hour daily coverage.
6. RBI-Compliant Export Invoicing Got Smoother
GST-compliant export-of-services invoicing (LUT/LOA) is now standard for serious Indian agencies. US/UK/Gulf clients pay USD/GBP/AED directly to an Indian agency bank account; the agency handles all GST and income tax. No 18% GST is charged to foreign clients on export of services.
7. Equity Compensation Becoming Common
Indian senior engineers increasingly accept partial equity for funded startups. 0.1-0.5% ESOPs + ₹1L-₹2L cash is becoming common for Series-A startups hiring senior India engineers.
8. MongoDB Atlas Is Now Default
Self-hosted MongoDB is increasingly rare in 2026. MongoDB Atlas (managed) is the default for new builds — meaning your MERN engineer needs Atlas-specific knowledge (Atlas Search, Atlas Vector Search, Atlas Triggers, network peering).
Anonymized Client Stories
Story 1: Bengaluru-Based D2C Founder — Saved ₹38L by Going MERN Squad Over US Agency
A Bengaluru-based D2C personal-care founder we worked with had been quoted ₹70L by a US agency to rebuild her Shopify-plus-custom-backend ecosystem. She came to us through a referral. We placed a MERN squad — 1 senior, 2 mid-level engineers + a part-time PM — at ₹4.8L/mo total. Project shipped in 6 months for a total of roughly ₹32L. She kept the squad on at ₹3.2L/mo for ongoing roadmap work and went on to scale to 8-figure annual revenue without rebuilding the stack.
Savings: roughly ₹38L vs the US agency quote, plus continuous capacity afterward.
Story 2: UK SaaS Founder — Burned by Cheap Freelancer, Then Hired Right
A UK fintech SaaS founder hired a ₹30K/mo "senior" MERN developer directly via LinkedIn. Developer ghosted at month 3 with critical payment processing code unfinished. The founder lost roughly ₹3.5L + 14 weeks of timeline + a missed launch window.
He came to us. We placed a mid-senior MERN engineer at ₹1.4L/mo. The engineer rebuilt the payment flow in 5 weeks for a fixed ₹3L scope, then continued on a monthly retainer. The founder said the difference in quality justified more than 4x the cost.
Story 3: Dubai-Based Marketplace Startup — Embedded MERN Engineer
A Dubai-based B2B marketplace startup with a 3-person local team needed to add backend capacity without slow Dubai hiring (where senior MERN engineers run ₹6L-₹8.5L/mo). They embedded an Indian senior MERN engineer in their team at ₹1.8L/mo. With only 1.5 hours of time-zone difference, the engineer joined daily Dubai standups easily and worked normal hours.
Result: shipped 6 months of backlog features in 4 months. Hired 2 more Indian MERN engineers on the same embedded model. The entire backend team is now India-based, and the Dubai team focuses on sales + product.
Story 4: Mumbai-Based B2B SaaS — Full-Time vs Agency Decision
A Mumbai-based B2B SaaS that already had 3 in-house engineers wanted to add 2 more MERN developers. They considered full-time via Naukri (estimated ₹18L/year per engineer all-in = ₹3L/mo each) vs Codingclave's dedicated engineers at ₹1.6L/mo each.
They picked us for a 6-month bridge while they figured out long-term hiring. Six months in, they extended the contract because: (1) bench depth meant zero stoppage when one engineer fell sick for 10 days, (2) GST-compliant invoicing was simpler than payroll for 5 engineers, (3) the agency PM handled performance management.
How Codingclave Hires Right for You
Our placement process for MERN stack engineers:
| Day | What Happens |
|---|---|
| Day 1 | Discovery call (60 min) — your needs, stack, timeline, budget |
| Day 2-3 | We propose 2-3 vetted MERN candidates with portfolios, GitHub links, short video intros |
| Day 3-5 | You interview, run technical screening, pick one |
| Day 5-7 | Contract, onboarding, first commit |
| Week 2 | First feature shipped, weekly cadence established |
| Week 4 | Trial-period check-in — keep, swap, or scale |
Every engineer is vetted on 4 layers before we propose them: GitHub + portfolio review, written async test, live coding, and a paid trial week with us internally.
Pricing Tiers — Codingclave MERN Stack Hiring
| Tier | What's Included | Monthly | Best For |
|---|---|---|---|
| Solo MERN Engineer | 1 vetted mid-senior engineer, async communication, weekly demos | ₹1.4L-₹2.4L | Startups adding engineering capacity |
| MERN Pair | 1 senior + 1 mid-level engineer, joint code reviews, shared on-call | ₹2.8L-₹4.2L | Funded startups, faster delivery |
| MERN Squad | 1 senior + 2 mid + part-time PM + part-time QA | ₹4.8L-₹7L | Complex builds, full team needed |
| Embedded Engineer | Single engineer integrated into your team, joins your standups | ₹1.5L-₹2.8L | Existing teams adding capacity |
| Fixed-Price Project | Scoped MVP or feature, defined deliverables, milestone-based | ₹3L-₹25L | Founders with clear scope |
All tiers include: GST-compliant export invoicing, dedicated Slack channel, weekly demos, replacement guarantee, and IP assignment.
Related Codingclave Guides
- Hire Node.js Developer India 2026 — Real Pricing Guide
- Hire React Developer India 2026
- Hire Flutter Developer India 2026
- Next.js Role-Based Routing Guide 2026
- Build vs Buy CRM in India 2026
Who Wrote This Guide
I am Ashish Sharma — founder of Codingclave, a Top Rated Upwork agency from Lucknow, India. Over 8+ years I have placed 70+ MERN stack engineers with US, UK, Australian, Gulf, and Indian clients, and personally interviewed 600+ MERN candidates across the experience spectrum. We are a Top Rated Upwork agency, which means our engineers consistently hit the top quality bar across Upwork's full pool of India agencies.
If you are weighing a MERN stack hire — whether for an MVP, an ongoing build, or augmenting an existing team — message me directly on WhatsApp at +91 92771 84741. I respond personally within a few hours during India business hours, and I will tell you honestly whether Codingclave is the right fit or whether you should hire a different way.
Connect on LinkedIn — I share weekly notes on offshore hiring, MERN architecture, and what is actually working in 2026 SaaS builds.