AI-native software is a work in progress.We make it work at enterprise scale.
From specification and independent 3rd-party codebase audits to production deployment — tracking our complete portfolio of 0→1 ventures, multi-tenant platforms, and applied AI systems.
Production software systems across ideation, active development, market scale, and past founding experiences.
Active · ScaleAug 2025
Malook Doctor CRM & Clinical Microsites
Multi-Tenant Subdomains · Real-Time OPD Booking · WhatsApp Lead Sync
Multi-tenant clinical appointment CRM and high-converting patient microsite engine built with Next.js 16 (Turbopack), Prisma 7 ORM, PostgreSQL, and WhatsApp lead capture.
Active
Healthcare SaaS
Doctor CRM
Next.js 16
Prisma 7
PostgreSQL
WhatsApp API
Problem & Opportunity
Medical practices and specialist doctors often rely on fragmented appointment booking links, generic directory listings, or slow static sites that fail to capture patient inquiries or synchronize with clinic OPD schedules. Malook Doctor CRM provides doctors with high-converting, personalized clinical microsites integrated directly into an end-to-end appointment management dashboard.
Key Architecture
Next.js 16 & Prisma 7 ORM: Built on Next.js 16 App Router with Turbopack and Prisma 7 PostgreSQL driver adapter architecture for sub-100ms server rendering.
Multi-Tenant Subdomain Routing: Dynamic tenant routing (/t/[slug]) mapping doctor profiles, custom domain aliases, and OPD availability rules.
Lead Capture & WhatsApp Integration: Instant notification pipeline broadcasting new appointment inquiries directly to clinic staff and doctors via WhatsApp webhooks.
Rich Media & Content Engine: Modular CMS supporting clinical photo galleries, video interviews, press releases, patient FAQs, and verified Google reviews.
GTM & Traction
Powering flagship clinical microsites including Dr. Vibhor Upadhyay (Senior Consultant Neurologist, Apollo Hospitals Lucknow).
100% automated slot allocation for OPD consultations and emergency stroke/neurology inquiries.
Direct lead conversion increase through glassmorphic UI cards, verified patient review carousels, and responsive 1-click WhatsApp booking.
Ideation · R&DFeb 2025
Forge & Beacon: AI Evaluation Engine
Sub-Second Replay · Bounded Concurrency · CI Quality Gate
High-throughput Rust parallel eval runner (Forge) and Elixir Phoenix LiveView ops console (Beacon) built to block LLM regressions in production CI/CD.
Ideation
R&D
Rust
Elixir
AI Evals
Tokio
Phoenix
Problem & Opportunity
As enterprise teams deploy tool-using agents and RAG pipelines, prompt and model updates frequently introduce silent regressions in quality, accuracy, and permission bounds. Existing evaluation frameworks are slow and difficult to gate in fast-moving CI/CD pipelines.
Key Architecture
Forge (Rust): High-performance CLI eval runner using Tokio for bounded concurrency, deterministic replay, and CI exit codes.
Beacon (Elixir/Phoenix): Real-time LiveView dashboard managing OTP supervisors per evaluation job with live PubSub progress.
Active R&D and internal dogfooding across applied AI client projects.
In Development · BetaJan 2025
Travel Professor: Agentic Booking Concierge
Multi-Step Reasoning · Custom MCP Server · Live B2B Integration
Autonomous travel agent on Crux with multi-step reasoning for hotel discovery, price comparison, and custom itineraries via Model Context Protocol (MCP).
In Development
Beta
Agentic AI
MCP
LangChain
Python
pgvector
Problem & Opportunity
Standard travel booking engines require users to navigate hundreds of static filter combinations. Travel Professor bridges high-concurrency B2B inventory APIs directly to natural language LLMs, allowing users to plan, negotiate, and customize complex itineraries autonomously.
Key Architecture
Model Context Protocol (MCP): Custom MCP server exposing private B2B travel APIs securely to local and cloud LLMs.
Agentic Workflows: Multi-step reasoning loops using LangChain, Python, and pgvector for semantic accommodation retrieval.
Guardrails & Evaluation: Automated trace logging and CI-gated quality evaluations to prevent hallucinated pricing.
GTM & Traction
Currently in closed beta testing with selected enterprise travel partners.
Open-source core tools available on GitHub.
Active · ScaleOct 2024
Crux B2B API Layer
1M+ Daily Reqs · 50+ API Integrations · 99.9% Uptime
High-concurrency API aggregator integrating 50+ hotel, DMC, and payment providers. Handling 1M+ daily requests with sub-200ms hybrid search & RAG.
Active
Scale
Crux API
Go
Node.js
AWS
RAG
Problem & Opportunity
B2B travel providers operating legacy infrastructure experience fragmented API contracts, high latency, and frequent downstream downtime. Crux was built as a unified, high-concurrency API aggregation layer to normalize responses, route requests dynamically, and provide sub-200ms hybrid search.
Key Architecture
Built with Go microservices and Node.js/TypeScript event loops.
Hybrid Search & RAG: Vector embedding retrieval combined with structured PostgreSQL filtering for live inventory lookup.
Direct-to-consumer apparel venture built from ideation through full-stack execution, market validation, and beta customer acquisition.
Past
Founding
0→1 Venture
E-Commerce
Product
Full-Stack
Problem & Opportunity
Identified market fragmentation in custom apparel retail. Founded MinedeniM to own the end-to-end product lifecycle from brand strategy and supply chain prototyping to full-stack e-commerce execution.
Key Learnings & Takeaways
Mastered rapid 0→1 market validation and customer feedback loops.
Hands-on experience balancing technical debt against speed to market.
Foundation for future B2B SaaS product management and system engineering.
02 / CONSULTING & ADVISORY
What teams bring us.
You know the business. We bring the AI-native architecture, independent verification, and engineering rigor to ship critical systems without drift.
When software is built by an external agency, outsourced vendor, or departing technical team, leadership often lacks visibility into hidden security debt, brittle schemas, and scaling bottlenecks. We perform a line-by-line independent audit and deliver a board-ready technical report.
DIAGNOSTIC & REPORTSCOPED BY CODEBASE SPAN
$30USD / hour
Architecture Review & 3rd-Party Code Audit
Total hours are determined transparently after an initial check of your current codebase and architecture span (repositories, LOC, services, database schema, and cloud footprint). Culminates in a complete written Audit & Scale Report.
HANDS-ON IMPLEMENTATIONEXECUTION & FIXES
$20USD / hour
Code Execution & Post-Audit Remediation
Need us to fix the vulnerabilities, refactor bottlenecks, or build new features after the audit? Hands-on coding, PR implementation, test writing, and bug remediation are billed at our standard engineering rate.
TRANSPARENT COSTING MATRIX
How we scope audit hours by codebase span
Every engagement begins with a 1-hour span check to inspect your repository metrics and lock in a fixed hour ceiling before deep auditing starts.
STEP 01 · SPAN TRIAGE
Initial Codebase Span Check
1 Hour$30 USD
Live repo & architecture walkthrough to measure lines of code (LOC), service count, DB tables, and infra footprint. Outputs a locked hour quote for the full audit.
Repo structure & dependency tree scan
Database & cloud boundary inventory
Fixed hour & delivery commitment
SPAN TIER A · SMALL / MVP
Single-Repo / Agency Handover
4 – 6 Hours$120 – $180 USD
For early-stage MVPs, web/mobile frontends, or single-service agency builds prior to milestone sign-off or final vendor payment release.
Up to ~15,000 lines of code (1–2 repos)
Single database & authentication flow
Full 8-pillar written report in 48 hours
SPAN TIER B · PRODUCTION SAAS
Full-Stack SaaS, CRM & APIs
10 – 15 Hours$300 – $450 USD
For revenue-generating B2B SaaS, healthcare/clinical CRMs, multi-role admin portals, payment gateways, and third-party webhook integrations.
15,000 – 50,000 LOC (Frontend + API + Workers)
Multi-tenant DB schema & RBAC security review
Full 8-pillar report + 60-min executive readout
SPAN TIER C · ENTERPRISE & AI
Distributed Systems & AI Stack
20 – 30+ Hours$600 – $900+ USD
For high-concurrency microservices, M&A technical due diligence, AWS/Azure Terraform infrastructure, and production LLM/RAG/agentic pipelines.
Inside the 8-Pillar Complete Audit, Handover & Scale Report
Every finding is traced to exact file paths and line numbers, rated by severity (Critical / High / Medium / Low), and paired with concrete code diffs and a vendor sign-off checklist.
01 / SECURITY VULNERABILITIES
Security & Threat Surface
Authentication & session flaws, broken RBAC/tenant isolation, SQL/NoSQL/XSS injection vectors, hardcoded API keys in Git history, insecure webhooks, IAM over-permissions, and dependency CVEs.
02 / SOW & MOCK-CODE DETECTION
Contract vs. Reality Verification
Cross-checks your vendor’s Statement of Work (SOW) against actual Git commits. Uncovers "demo-ware" (hardcoded UI mocks pretending to be real backend logic), stubbed edge cases, and unreviewed AI-generated boilerplate.
03 / CODE PRACTICES & HYGIENE
Maintainability & Test Coverage
Identifies copy-pasted or dead logic, modularity anti-patterns, missing TypeScript/schema validation, silent error swallowing, and gaps in unit/integration/E2E test suites.
04 / PERFORMANCE & COST OPTIMISATIONS
Latency & Cloud Spend Teardown
N+1 database query detection, missing composite indexes, memory leaks, bundle/payload bloat, Redis/CDN caching gaps, and AWS/Azure infrastructure right-sizing to cut monthly burn by 30–50%.
05 / ARCHITECTURE COMMENTARY
System Design & Failure Resilience
Evaluates service boundaries, database normalization & migration safety, payment/webhook idempotency, background queue retry/dead-letter design, and zero-downtime deployment pipelines.
06 / VENDOR LOCK-IN & HANDOVER READINESS
"Bus-Factor" & IP Ownership Audit
Verifies whether a new engineer can boot the stack in <15 mins. Audits root ownership of cloud accounts, DNS, payment keys, Terraform state, environment secrets, and undocumented tribal steps.
07 / DATA PRIVACY, COMPLIANCE & BACKUPS
GDPR / HIPAA / DPDP & Disaster Recovery
Checks for PII or patient/financial data leaked in application logs, encryption at rest/in transit, immutable audit trails, automated DB point-in-time recovery, and RPO/RTO backup verification.
08 / AI & LLM GUARDRAILS (IF APPLICABLE)
Prompt Security & Token Economics
Audits prompt-injection exposure, denial-of-wallet / unbounded token spend (missing per-tenant rate & cost caps), PII leakage to model APIs, RAG retrieval accuracy, and fallback provider resilience.
Beyond technical commentary, you receive a board- and vendor-ready decision package: a Milestone Release Verdict (whether to approve or hold back final agency payout), a P0 Launch Blocker Checklist you can hand directly to your 3rd-party vendor to fix at their expense, and a 10×–100× Scale Blueprint with itemised remediation hours pre-priced at our $20/hr execution rate.
SAMPLE REPORT ARTIFACTVERDICT: HOLD MILESTONE PAYOUT
Start with a single hourly audit or coding block to see how we work, or commission a fixed-scope turnkey platform build with transparent milestone terms.
Currency
TRACK A // ON-DEMAND ENGINEERING & VERIFICATIONBilled hourly or in 5–10 hr blocks
01 / HANDS-ON ENGINEERINGEXECUTION RATE
Code & Implementation
$20/ hour (USD)
Hands-on feature development, post-audit vulnerability remediation, bug fixes, API integrations, and performance refactoring.
Production PRs & Feature Builds
Post-Audit P0/P1 Remediation
API, Webhook & DB Engineering
Unit & Integration Tests
02 / DIAGNOSTIC & REPORT3RD-PARTY CODE AUDIT
Architecture & Code Audit
$30/ hour (USD)
Independent audit of 3rd-party agency or internal codebases. Total hours locked after checking current code & architecture span.
Security & Auth Vulnerability Scan
Code Practices & Vendor Quality
DB, Latency & Cloud Cost Audit
8-Pillar Report & P0–P2 Scale Plan
TRACK B // TURNKEY PLATFORM & AI FACTORY BUILDSFixed project deliverables & milestone SLAs
TIER 01 / FOUNDATION
Website (Static)
High-performance static site for personal branding, clinical microsites, or product landing pages with instant load speeds and zero maintenance overhead.
Included Architecture
100% Static Jamstack Architecture
Responsive Mobile-First & Dark/Light UI
Technical SEO, Schema.org & Fast CDN
AWS S3 + CloudFront / EC2 Ready
Engagement Terms
100% upfront · 5-day turnaround · 1 revision round included
$200/ project (USD)
TIER 02 / FULL-STACK COREMOST DEPLOYED
Website+ (Website + CRM)
Complete web presence integrated with lead capture, patient/client appointment scheduling, and a customer relationship management (CRM) backend.
Included Architecture
Everything in Static Website Tier
Integrated Lead & Appointment CRM
Automated Contact & Inquiry Intake
Admin Control Panel & Records DB
Engagement Terms
50% initial deposit / 50% QA sign-off · 14-day turnaround · 30 days support
$600/ project (USD)
TIER 03 / INTEGRATED SUITE
Website+ & Plugins
Enterprise-ready web & CRM platform integrated with third-party service plugins: WhatsApp Business API, Google Workspace/Maps, and Payment Gateways.
Included Architecture
Full Website + CRM Portal Stack
WhatsApp Business Webhooks & Alerts
Google Calendar Sync & Maps API
Razorpay / Stripe Gateway Setup
Engagement Terms
40% deposit / 30% milestone / 30% launch · 21-day build · 60 days SLA support
$1,200/ project (USD)
TIER 04 / GO BIG · ENTERPRISE
AI-Native Software Factory
Bespoke Agentic AI architectures, custom Model Context Protocol (MCP) servers, RAG vector search, DSPy compilation, & Fractional CTO leadership.
Included Architecture
Custom MCP Servers & LLM Tooling
Sub-200ms RAG & Vector Search
CI-Gated AI Evals (PromptFoo, RAGAS)
Fractional CTO & Tech Due Diligence
Engagement Terms
Custom SOW following technical discovery call · Retainer or milestone-based
An SDLC discipline that refines business intent, governs agentic execution, verifies outputs, and preserves a complete architectural audit trail.
01
Requirements & Intent
Capture business constraints, latency bounds, and cost targets in plain English so everyone agrees on what is built before writing code.
02
Blueprints & Thin Vertical
Turn business intent into signed-off technical schemas, API contracts, and a working vertical prototype within weeks.
03
Work Orders & Execution
Decompose builds into traced tasks executed by senior engineers alongside MCP-connected AI coding agents.
04
Validation & Eval Gates
Enforce automated drift detection, CI/CD test suites, cloud cost guardrails, and LLM evaluation benchmarks prior to release.
06 / SCHEDULE AUDIT OR ADVISORY
Bring us the gnarly problem.
Reserve a 1-hour codebase span triage for a 3rd-party audit ($30/hr), book hands-on engineering time ($20/hr), or schedule a 0→1 discovery call.
3RD-PARTY AUDIT WORKFLOW
1. Grant read-only Git/cloud access or screen-share during your 1-hr Span Triage ($30). 2. Receive a locked hour quote based on codebase span. 3. Receive the 5-pillar written Audit & Scale Report + walkthrough.
07 / CONTACT & RETAINERS
Talk to us.
Available for independent 3rd-party codebase audits, selective B2B SaaS builds, AI-native control planes, and venture partnerships.