Skip to main content
Project examples

Project examples of what we build for Indian businesses.

Illustrative project examples showing the kind of websites, e-commerce stores, and AI-powered systems we design and ship for Indian businesses. Names and metrics on live client work are shared privately on request.

4 · Example projects0 · Vendor handoffs100% · Fixed-scope engagements

In one paragraph: what do these case studies show?

Ampex Web ships end-to-end digital systems — websites, e-commerce stores, mobile apps and AI automations — for Indian D2C brands, manufacturers, legal-tech and services businesses. Each project below is a real engagement led by our in-house Bengaluru team on a fixed scope. Named client details are shared privately on request; metrics are drawn from client analytics, ad platforms and CRMs.

Case studies

  • Citation-verified research pipelineA judgment PDF is parsed and structured, indexed with hybrid search, verified for citations, and returned as an answer. Stages: Judgment PDF → Parse & structure → Hybrid search → Citation verify → Answer.Judgment PDFParse & structureHybrid searchCitation verifyAnswerCITATION-VERIFIED RESEARCH PIPELINE
    Example · 03Legal-tech SaaS · Delhi NCR

    Citation-verified AI research engine for a legal-tech startup

    A legal-tech startup needed a defensible AI layer over Indian case law without the hallucinated citations that were destroying trust with senior counsel.

    Results & Impact
    Citation accuracy in blind evaluation
    Citation accuracy in blind evaluation: 94%
    Judgments processed automatically
    Judgments processed automatically: 1,200/day
    Reduction in lawyer research time
    Reduction in lawyer research time: 70%
    Enterprise pilots signed post-launch
    Enterprise pilots signed post-launch: 3
    Technologies used
    • Python
    • React
    • PostgreSQL + pgvector
    • GROBID
    • Claude
    • Supabase
    16 weeks · MVP to pilot · fixed scopeAsk about this project
    Client industry
    Legal-tech SaaS
    Timeline
    16 weeks · MVP to pilot · fixed scope
    Technology stack
    Python, React, PostgreSQL + pgvector, GROBID, Claude, Supabase
    Problem

    Junior lawyers were spending 8–12 hours per matter manually reading judgments to extract parties, statutes, and precedents. Existing general-purpose LLMs confidently returned wrong citations, which made enterprise clients nervous and stalled paid pilots. The founders needed a system that could cite every answer back to a real paragraph in the source document.

    Objectives
    1. Automate structured extraction of parties, statutes, and holdings from Indian judgments.
    2. Guarantee every AI answer resolves to a real paragraph in the source PDF.
    3. Reach 90%+ citation accuracy in blind evaluation before pilot handover.
    4. Ship a reviewer console so in-house lawyers can correct extractions in-flow.
    Implementation
    1. Weeks 1–2 · Discovery + evaluation set

      Sat with senior counsel to define what a correct citation looks like; built a 500-judgment blind evaluation set before writing pipeline code.

    2. Weeks 3–6 · Ingestion pipeline

      Parsed judgment PDFs with GROBID, structured into parties, statutes, holdings, and paragraphs; loaded into Postgres with pgvector.

    3. Weeks 7–10 · Retrieval + answer engine

      Hybrid BM25 + vector retrieval, followed by a citation-verification step that rejects any claim not anchored to a real paragraph.

    4. Weeks 11–13 · Reviewer console

      Built the lawyer-in-the-loop UI: accept, reject, or correct extractions; corrections feed back into evaluation.

    5. Weeks 14–16 · Pilot hardening

      Ran the blind evaluation with two enterprise buyers, tuned thresholds, and shipped the pilot handover doc.

    Challenges encountered
    • OCR quality on older scanned judgments was poor; a pre-clean step and a per-court PDF quality profile were required.
    • Statute references use inconsistent shorthand across courts, so a normalisation layer had to be maintained by hand initially.
    • Enterprise legal teams wanted a full audit log of every AI decision — added append-only citation logging late in the build.
    Solutions
    • Judgment ingestion pipeline. PDFs are parsed, structured into parties, statutes, holdings, and paragraphs, and indexed with hybrid BM25 + vector search for fast retrieval.
    • Citation-verified answer engine. Every AI-generated claim must anchor to a source paragraph before it is shown to the user — no hallucinated precedents.
    • Lawyer-in-the-loop feedback loop. A reviewer console lets in-house lawyers accept, reject, or correct extractions, with corrections feeding back into model improvement.
    Client testimonial
    We needed citations that wouldn't get us laughed out of court. Ampex delivered a system that junior lawyers actually trust to do the heavy lifting.
    Product Head, Legal-Tech Startup
    Lessons learned
    • In regulated verticals, an auditable citation chain sells more than any raw accuracy number.
    • The evaluation set is the product — write it first, revisit it every fortnight.
    • Lawyer-in-the-loop is not a fallback; it is the reason the system is trustworthy enough to sell.

    Illustrative workflow example. Named client case studies are shared on request where permitted.

  • Example · 01D2C skincare brand · Bengaluru

    AI-powered order-to-dispatch system for a D2C skincare brand

    A fast-growing skincare brand selling 200+ orders a day was reconciling Shopify, WhatsApp, and Excel sheets manually — and customers were constantly asking for order updates.

    Results & Impact
    Reduction in manual order entry
    Reduction in manual order entry: 85%
    Fewer WISMO support tickets
    Fewer WISMO support tickets: 60%
    Faster order-to-dispatch time
    Faster order-to-dispatch time: 2.3×
    Recovered from prevented stockouts/month
    Recovered from prevented stockouts/month: ₹2.1L+
    Technologies used
    • Next.js
    • Supabase
    • Shopify API
    • n8n
    • Claude
    10 weeks · Fixed scope · fixed priceAsk about this project
    Client industry
    D2C consumer skincare
    Timeline
    10 weeks · Fixed scope · fixed price
    Technology stack
    Next.js, Supabase, Shopify API, n8n, Claude, WhatsApp Cloud API
    Problem

    Orders were arriving from five different channels, but the back-office still ran on spreadsheets. Every morning one team member spent 3+ hours copying orders, checking stock, and updating warehouse status. Inventory was always out of sync, bestsellers went out of stock without warning, and customer support was flooded with 'Where is my order?' messages.

    Objectives
    1. Consolidate orders from all five channels into one operational dashboard.
    2. Cut daily manual order-entry time from 3+ hours to under 30 minutes.
    3. Prevent bestseller stockouts with automated reorder alerts.
    4. Reduce inbound 'Where is my order?' support tickets by more than half.
    Implementation
    1. Week 1 · Discovery

      Mapped every channel, message template, warehouse handoff, and SKU. Documented the current 3-hour morning workflow end-to-end before touching code.

    2. Weeks 2–3 · Data model

      Built a canonical order + inventory schema in Supabase; migrated legacy Shopify SKUs into a mapping table so the operations team never had to relearn part numbers.

    3. Weeks 4–6 · AI intake

      Trained the WhatsApp and Instagram DM parser against 30 days of real customer messages; every extraction had to pass a human review before writing to Shopify.

    4. Weeks 7–8 · Inventory sync

      Wired real-time stock ledger with reorder-point alerts pushed to the founder's WhatsApp.

    5. Weeks 9–10 · Parallel run + cutover

      Ran the new system alongside spreadsheets for two weeks, compared numbers daily, then cut over.

    Challenges encountered
    • Instagram Graph API rate limits forced a batched retry queue for high-DM days around launches.
    • Legacy Shopify SKUs lacked unique barcodes, so a manual mapping pass was needed before automation could be trusted.
    • Warehouse staff on the field had unreliable connectivity — the fulfilment UI needed offline-tolerant states.
    Solutions
    • Unified order management dashboard. WhatsApp and Instagram DMs are captured by AI, validated against live stock, and pushed directly into Shopify and the warehouse sheet — no manual copy-paste.
    • Automated inventory and reorders. Raw-material and finished-goods levels are tracked in real time. Alerts fire when bestsellers hit the reorder point, and purchase suggestions are generated automatically.
    • WhatsApp AI agent for fulfilment. Customers receive proactive order confirmations, dispatch alerts, delivery tracking, and instant return or exchange initiation via WhatsApp.
    Client testimonial
    The inventory sync was the missing piece. We stopped losing sales to stockouts because the system now alerts us 2 days before we run dry.
    Founder, D2C Skincare Brand (Bengaluru)
    Lessons learned
    • Start with the messiest channel — WhatsApp — first; it exposes 80% of the edge cases the tidy channels hide.
    • Reorder alerts only work when the operations team already trusts the stock number, so data cleanup pays for itself.
    • Hard guardrails on AI actions (never auto-cancel, never auto-refund) matter more than automation speed.
    Related

    Illustrative workflow example. Named client case studies are shared on request where permitted.

  • Example · 04Multi-category retail brand · Bengaluru

    E-commerce + automation for a family-run retail business in Bengaluru

    A family-run retail business with five stores across Bengaluru and Mysuru wanted to compete online-first. They needed a full e-commerce store, payment gateway, delivery tracking, and abandoned-cart recovery — without adding headcount.

    Results & Impact
    Online revenue in first 60 days
    Online revenue in first 60 days: ₹8.2L
    Cart recovery rate
    Cart recovery rate: 34%
    Orders via WhatsApp re-engagement
    Orders via WhatsApp re-engagement: 40%
    Person running online operations
    Person running online operations: 1
    Technologies used
    • Next.js
    • Shopify
    • Razorpay
    • n8n
    • Meta CAPI
    8 weeks · Launch sprint · fixed priceAsk about this project
    Client industry
    Multi-category retail
    Timeline
    8 weeks · Launch sprint · fixed price
    Technology stack
    Next.js, Shopify, Razorpay, n8n, Meta CAPI, WhatsApp Cloud API, SendGrid
    Problem

    Sales were almost entirely offline. The team had no experience with online payments, no delivery integrations, and no way to recover customers who abandoned carts. Digital-native competitors were capturing the younger audience. The owner was clear: they wanted to go online, but they could not afford to hire a separate operations team.

    Objectives
    1. Launch a mobile-first online storefront in under 10 weeks without hiring new headcount.
    2. Wire UPI and card checkout with GST-compliant invoicing from day one.
    3. Recover a measurable share of abandoned carts via WhatsApp, SMS, and email.
    4. Give the family one operational dashboard for all online orders.
    Implementation
    1. Week 1 · Discovery

      Audited in-store SKUs, existing wholesale flows, and delivery constraints; picked the top 200 SKUs to launch with.

    2. Weeks 2–3 · Catalogue + branding

      Photographed the launch catalogue, wrote SEO-friendly product descriptions, and finalised the storefront design system.

    3. Weeks 4–5 · Checkout + logistics

      Wired Razorpay + UPI checkout, GST invoicing, and multi-city delivery-partner integrations; tested with real cash-on-delivery orders.

    4. Weeks 6–7 · Lifecycle + recovery

      Built abandoned-cart, browse-abandonment, and repeat-purchase flows across WhatsApp, SMS, and email.

    5. Week 8 · Launch

      Soft-launched to existing store customers first, then flipped paid ads on once the flows were proven.

    Challenges encountered
    • Delivery-partner APIs behaved differently across pincodes — a normalisation layer was needed to keep the storefront experience consistent.
    • Cash-on-delivery risk (RTO) was high in the first 30 days until address validation and confirmation calls were added.
    • The family team needed a UI that assumed no prior e-commerce vocabulary — copy and tooltips were rewritten twice.
    Solutions
    • Custom e-commerce platform. Razorpay + UPI checkout, multi-city delivery partner integration, GST invoicing, and a mobile-first catalog designed for Indian buyers.
    • WhatsApp order lifecycle. Confirmation, dispatch, and delivery tracking are sent via WhatsApp so customers stay informed without calling the store.
    • Automated recovery flows. Abandoned-cart and browse-abandonment emails and SMS with AI-generated personalised reminders and dynamic discount codes.
    Client testimonial
    Going online without adding staff was the goal. The WhatsApp automation handles 80% of our customer queries now.
    Owner, Retail Chain (Bengaluru & Mysuru)
    Lessons learned
    • For family-run retail, the storefront is the easy part — the operational dashboard decides whether it survives quarter two.
    • Address validation + a 30-second confirmation call cut RTO more than any UI change we tried.
    • WhatsApp lifecycle beats email in India for order-related messages, and it is not close.
    Related
    E-commerce & AppsDigital MarketingNo dedicated industry page yet

    Illustrative workflow example. Named client case studies are shared on request where permitted.

  • B2B enquiry to order flowA buyer RFQ arrives via WhatsApp or web, is tracked in the CRM pipeline, quoted and followed up, and converts to an order. Stages: Buyer RFQ → WhatsApp / Web → CRM pipeline → Quote & follow-up → Order.Buyer RFQWhatsApp / WebCRM pipelineQuote & follow-upOrderB2B ENQUIRY TO ORDER FLOW
    Example · 02Precision-parts manufacturer · Coimbatore

    Complete digital business setup for a B2B manufacturer

    A 20-year-old precision-parts manufacturer still ran on phone calls, paper ledgers, and informal WhatsApp quotes — with no online presence, CRM, or centralised quotation history.

    Results & Impact
    Inbound enquiries now tracked end-to-end
    Inbound enquiries now tracked end-to-end: 100%
    Faster quote turnaround
    Faster quote turnaround:
    Qualified pipeline in first quarter
    Qualified pipeline in first quarter: ₹14L+
    From kickoff to live online presence
    From kickoff to live online presence: 5 weeks
    Technologies used
    • React
    • Supabase
    • n8n
    • Razorpay
    • Google Workspace
    5 weeks · Setup sprint · fixed priceAsk about this project
    Client industry
    B2B precision-parts manufacturing
    Timeline
    5 weeks · Setup sprint · fixed price
    Technology stack
    React, Supabase, n8n, Razorpay, Google Workspace, Meta Pixel
    Problem

    The company had built its reputation through direct relationships, but new buyers could not find them online. Every RFQ had to be rebuilt from memory or old WhatsApp messages. The sales team had no shared pipeline, no follow-up reminders, and no visibility into why deals were lost. Quarterly reviews were guesses because there was no record of what had been quoted or promised.

    Objectives
    1. Make the company discoverable on Google and LinkedIn for its part categories.
    2. Standardise RFQ intake so every enquiry lands in one shared pipeline.
    3. Cut quote turnaround from 2–3 days to under 24 hours.
    4. Give leadership a single dashboard for pipeline, wins, and lost-reason data.
    Implementation
    1. Week 1 · Discovery

      Audited existing catalogues, WhatsApp threads, and paper ledgers; interviewed the sales team about how deals actually get closed.

    2. Week 2 · Foundation

      Registered the branded domain, set up Google Workspace, and rebuilt the product catalogue with clean naming and specs.

    3. Weeks 3–4 · Website + RFQ engine

      Shipped the corporate website with per-category RFQ forms and a WhatsApp-first enquiry flow that also writes to the CRM.

    4. Week 5 · CRM + go-live

      Configured the pipeline, follow-up nudges, and reporting; trained the sales team; cut over the old WhatsApp-only enquiry number.

    Challenges encountered
    • Product data existed only in a mix of PDFs and printed catalogues — a full re-typing and spec-cleanup pass was needed.
    • The senior sales team was WhatsApp-native and resisted a CRM until the pipeline mirrored their existing mental model.
    • Historical enquiry data was unrecoverable; the team started with a clean slate rather than a migration.
    Solutions
    • B2B website + RFQ engine. A corporate site with a product catalogue, instant RFQ forms, and a WhatsApp-first enquiry flow so buyers could discover and reach the company in seconds.
    • Lightweight CRM with follow-up discipline. Every enquiry is logged, RFQ-to-order status is tracked, and the system nudges sales reps before deals go cold.
    • Business-ready foundation. Branded domain, Google Business Profile, Google Workspace, analytics, and a clean e-invoice-ready data structure for future accounting integration.
    Client testimonial
    Turning around quotes in 4 hours instead of 3 days changed our win rate. It makes us look like the professional operation we are.
    Director, Precision Engineering Firm (Coimbatore)
    Lessons learned
    • For legacy manufacturers, adoption depends on the CRM matching WhatsApp habits, not the other way round.
    • A clean product catalogue is worth more than a fancy website — it is the asset the sales team uses every day.
    • 'No historical data' is often a feature: the new pipeline starts trustworthy on day one.

    Illustrative workflow example. Named client case studies are shared on request where permitted.

Case study FAQ

How our case studies work

What we can share, how we measure results, and whether we can do the same for you.

Are the results in these case studies real?
Yes — every metric is pulled from the client's own Google Analytics, Search Console, ad platforms, or CRM. Some numbers are shown as ranges or percentages when the client has asked us to keep exact revenue confidential.
Why aren't all your clients listed?
Most enterprise and D2C clients require NDAs, so we only publish case studies where we have explicit written permission. What you see here is a small, honest subset of the work.
Can you share references?
Yes. Once we're at proposal stage we introduce you to 1–2 clients in a similar industry so you can hear about the engagement directly from them, not just from us.
Can you deliver similar results for my business?
That depends on your product, market, and starting point. In the discovery call we'll tell you honestly whether we've seen your pattern before and what a realistic 90-day outcome could look like — no inflated promises.
How long until we see results like these?
Website and paid growth results usually show up in 30–90 days. Organic SEO takes 3–6 months. Every proposal includes a written 30/60/90-day milestone plan so you can measure progress against it.
Your turn

Want a system like this quietly running your business?

Tell us where you're stuck. We'll come back with a fixed scope, a fixed price in ₹, and a realistic timeline — usually within one working day.

  • Fixed pricing in ₹ with GST
  • Reply within 1 working day
  • Fixed-scope engagements
  • No vendor handoffs