■ AI Chatbot · RAG · Agent Development
RAG isn't "embed the docs and chat." Real RAG is engineering at every step.
Demo conversation · anonymised
grounded · cited
How do I configure SSO with our enterprise tier?
SSO is available on Enterprise. Configure it under Settings → Identity. SAML is supported with Okta, Azure AD, OneLogin.
Does it work with Okta SCIM?
Yes — SCIM is available via the Okta integration. Provisioning flows are documented under Settings → SCIM.
■ Talk to the lead architect
KB-grounded chatbot, in-product copilot, or an agent workflow for ops? Bring it directly to Ritesh — no SDR, no qualification call.
Ritesh Agarwal · Lead Architect, Appycodes · LinkedIn
■ The RAG pipeline
Six stages. Every one is a place to ship quality or skip it.
01
Chunking strategy
How you split the source documents matters more than which model you embed with. Sentence overlap, header preservation, code-block handling.
02
Embedding model choice
OpenAI text-embedding-3-small vs Voyage vs Cohere — cost-vs-quality tradeoffs at scale. We model the bill before we lock the schema.
03
Hybrid retrieval
Vector + BM25 keyword. Pure vector misses exact-match queries (SKU codes, error codes, proper nouns). Hybrid is the default.
04
Reranking
Second-pass model that orders top-K results before the LLM sees them. Sharp quality jump for marginal cost.
05
Grounding & citations
Answers cite source chunks. User verifies, system is auditable. No grounding = no production deployment.
06
Eval
Without a regression suite, every model swap is a guess. We ship eval before we ship the chatbot.
■ What you get
Six engagement shapes.
01
RAG chatbots
Over customer KBs, product docs, internal wikis. Proper chunking, hybrid retrieval, reranking, citations. Eval suite included.
02
Customer-support chatbots
Integrated with Intercom / Zendesk / Help Scout / custom. Hand-off to human agent when confidence falls below threshold.
03
In-product AI assistants
Copilots that operate inside the app's own data, not generic web search. The assistant knows your tenants, your data, your permissions.
04
Agent workflows (where it fits)
Research summarisation, lead enrichment, content drafting with approval gates. Not payments. Not irreversible actions. We draw the line.
05
Model selection & cost modelling
OpenAI / Anthropic / Gemini / open-weights. Realistic monthly cost projections before the architecture is locked.
06
On-device AI for mobile
CoreML / MLKit where latency or privacy demand it. Pairs with React Native engagements.
▼ RAG Quality vs Cost Curve
Where to spend the next dollar on a RAG system.
Monthly cost on the x-axis, answer quality (eval score) on the y-axis. The sweet spot is where the curve bends — hybrid retrieval + reranking is where most teams get the biggest quality lift per dollar.
Sweet spot ($500/mo · 72 quality): hybrid retrieval + reranking. Past this, you spend a lot more for a little more.
■ Evidence
Production AI integrations and the authority to teach the stack.
Named portfolio
Indian AI client alongside Rajasthan Royals, CultFit, and KhataBook
We build production AI integrations for clients in this tier — RAG over product surfaces, evaluation harnesses, model selection with cost projections. Engagement details disclosed under NDA.
Authority signal
We train professionals on the stack we ship with
Live training sessions for practising Chartered Accountants on AI-assisted app building. Stack taught: Lovable + Supabase + Vercel + Cloudflare + Make.com. We aren't just integrating AI; we're teaching it to professional audiences.
■ An honest framing on agents
Agents (autonomous tool-using AI loops) are exciting and unreliable. We build them where the use case tolerates non-determinism — internal-tool automations, research summarisers, draft generators with human approval. We don't build them where reliability matters more than autonomy: payments, irreversible actions, compliance-sensitive workflows. Buyers respect that line.
Pairs with AI SaaS product development (for consumer / B2B AI products), AI app completion (Lovable / Bolt graduation), Supabase (pgvector is a common RAG store), workflow automation (AI-in-the-loop scenarios), and API & integration for tying the chatbot to ticket systems.
■ Good fit if
- SaaS products with a KB / docs estate and a defined support-volume problem
- Teams who have prototyped a chatbot and need the production layer — grounding, eval, monitoring
- Operators who want in-product AI without re-architecting the rest of the application
- Buyers comfortable being told an agent isn't yet the right tool for their use case
■ Probably not a fit
- Hobby chatbots — start with ChatGPT custom GPTs
- Use cases that need deterministic guarantees (payments, compliance-sensitive workflows)
- Teams expecting AI to write itself — we don't ship without eval and grounding
■ Stack we ship on
Models, vectors, retrieval, orchestration, eval.
■ If you have docs and a support volume problem
Bring the corpus, the eval questions, and a monthly budget. We'll cost it before we build it.
■ Other services
We build a lot more than this.
Explore the rest of what we build.
SaaS Web App Development
MVP to production builds, multi-tenant, billing, AI features.
Learn moreReact Native App Development
iOS + Android in one codebase. Push, offline, OTA updates.
Learn moreWeb App to Native Mobile App
Convert your Lovable / Claude / Bolt web app to React Native or Flutter — backend reused.
Learn moreTechnical SEO for SaaS
Prerender, schema, Core Web Vitals — engineering-led SEO.
Learn moreCustom WordPress Development
B2B marketplaces, membership sites, headless WordPress.
Learn moreShopify Development Services
Custom themes, migration to Shopify, Shopify apps, supplier-feed automation.
Learn moreCustom WooCommerce Development
B2B wholesale, high-SKU catalogues, WooCommerce Subscriptions, Shopify-to-Woo migrations.
Learn moreMaintenance & Support
Post-launch stability, security, monthly improvements.
Learn moreAI App Completion
Take an AI-built prototype to a production-ready product.
Learn moreAI Prototype to Native App
Convert AI-built web prototypes into native mobile apps.
Learn moreApp Store Launch
Submit, configure and optimise for App Store + Play Store.
Learn moreTech Stack Migration
Modernise legacy systems with zero-downtime migrations.
Learn moreWhite Label Development
Engineering capacity for agencies, under your brand.
Learn moreStartup Launch Support
From idea to live product — design, build, launch, growth.
Learn moreAPI & Integration
Custom REST/GraphQL APIs and third-party integrations.
Learn moreAI SaaS Product Development
Multi-tenant AI SaaS with subscriptions and admin dashboards.
Learn moreB2B Marketplace Development
Marketplaces with credit, KYC, and trust infrastructure baked in.
Learn moreCustom LMS Development
Customised Moodle deployments and bespoke Moodle mobile apps.
Learn moreMembership & WooCommerce Subscriptions
Subscription recovery, dunning, migrations, combinatorial billing matrices.
Learn moreShopify Migration
Magento, WordPress, BigCommerce → Shopify. Redirect-engineering first.
Learn moreWordPress Security & Malware Removal
Forensic incident response — Magecart, backdoors, DDoS, GDPR Art. 33.
Learn moreLaravel Development & Modernisation
Four production Laravel platforms. ERPs, contracts, tendering.
Learn moreStripe Billing Engineering
Webhooks, metering, proration, dunning, multi-currency, reconciliation.
Learn moreCloudflare Edge Engineering
Workers, R2, WAF, Bulk Redirects. The full surface, not just the orange cloud.
Learn moreInternal Tools & Admin Dashboards
Tender automation, contract operations, compliance calendars, calculators.
Learn moreSupabase Development
RLS, multi-role auth, production-hardening Lovable / Bolt / v0 prototypes.
Learn moreHeadless WordPress & WooCommerce
WP + Next.js with preview, ISR, auth handoff, media pipeline, search.
Learn moreSanity CMS Development
Schema-first builds, multilingual setups, migrations from WP / Contentful / Strapi.
Learn moreVue.js & Nuxt Development
Four production Vue platforms. Vue + Laravel as the default pairing.
Learn moreWordPress Performance Optimisation
nginx, PHP-FPM, slow-query analysis, plugin audit. Diagnostic engineering, not plugin installs.
Learn moreCustom WordPress Plugin Development
Surgical plugin work — namespaced, idempotent, version-controlled. Not functions.php.
Learn moreKnowledge Base & Community Platforms
KB + forum + adjacent surfaces. Search, version-aware routing, role-based visibility.
Learn moreEvent Ticketing Platforms
Multi-day festivals, white-label venue groups, offline-capable door-staff apps.
Learn moreWorkflow Automation
Make.com, n8n, Zapier, and custom Node workers — chosen by volume.
Learn moreProgrammatic SEO Engineering
Calculator- and location-driven SEO surfaces with real data and SSR.
Learn more