Founding AI Engineer at XXAVI AI Tech

I build AI products end to end, and the systems that keep them running.

Backend, apps and infrastructure — built so they stay reliable, safe and cheap to run. Right now that's Prajna, an AI storybook app for kids. Before that, two SaaS products I shipped solo, with paying users.

$0.48 → $0.11
AI cost per story, Prajna
3,600+
automated tests, 3 codebases
< 150 ms
p95 & p99 API latency
40+
paying users, zero ad spend
01 — Selected work

Prajna

In Google Play review

AI storybook and creativity app for kids aged 3–12. Built as founding engineer at XXAVI AI Tech: backend, admin panel and Android app.

Problem

AI image generation is slow (30 seconds to 10 minutes), expensive and unreliable — and everything shown to a child has to be safe.

What I built

A Postgres-backed job queue (FOR UPDATE SKIP LOCKED behind an outbox dispatcher), so no paid job is lost between the database and the queue. Every job saves after each step, so a crash or deploy resumes instead of paying again. Failover across 3 AI providers, and a credit ledger that can't double-charge or double-refund.

Cost per 7-image story$0.11 — down 77% from $0.48
AI pipelines9, on one resumable engine
ProvidersOpenRouter, Google, OpenAI — automatic failover
API surface130+ typed endpoints · 48 tables
Tests3,600+ across backend, app and admin

TypeScript / Node.js / Express / oRPC / Zod / Drizzle / PostgreSQL / Valkey / AWS / React Native / Expo / RevenueCat

AttemptIQ

Live

UPSC diagnostic analytics. Turns mock test results into readiness insights: decaying topics, trap patterns, real readiness.

Paying users15+, zero paid marketing
Organic pageviews12,000+
Latencyp95 & p99 < 150 ms
Data model23 tables · 5 materialized views

Next.js / PostgreSQL / Redis / BullMQ / pgvector / Gemini

ScoreResume

Live

AI resume scoring against a specific job description, with an async pipeline so LLM timeouts never fail a request.

Paying users25+, zero paid marketing
Free users300+
Latencyp95 & p99 < 150 ms
Organic pageviews18,000+

Next.js / PostgreSQL / Redis / BullMQ / pgvector / LangChain

02 — Experience
Jul 2026 — Now

Founding AI Engineer · XXAVI AI Tech

I own the whole Prajna platform: backend, admin panel and Android app.

  • Crash-resumable AI engine running 9 pipelines. A crash or deploy resumes from the last finished step instead of paying again.
  • Cut AI cost 77% ($0.48 → $0.11 per story) by measuring cost per step and switching image models.
  • Failover across 3 AI providers with circuit breakers, timeouts and a hard spend cap. An AI judge checks every image before a child sees it.
  • Race-safe credit ledger with RevenueCat billing: one-statement charges, refunds that can't double-pay, 3 silent revenue leaks caught.
  • Typed API contract (oRPC + Zod, 130+ endpoints) as a private npm package — breaking changes fail at compile time.
Jan 2024 — Jun 2026

Full Stack Developer · TheBrandwick

Led engineering for four international clients across EU, UK and US markets — sprint planning, client communication and delivery.

Show 4 highlights +
  • Booking and FAQ systems for quoka.co.uk, with 90+ Web Vitals.
  • Led architecture for VastuWisdom and mentored 2 engineers.
  • AlexandersFA: design to deployment, 95+ SEO, retention up 35%.
  • Frontend for AllUnity, contributing to a 45% rise in conversion.
2023

Full-stack program · Masai School

2017

B.E. Mechanical Engineering · Don Bosco Institute of Technology

03 — About

Mechanical engineer by training. Software builder by choice.

I care about the unglamorous parts: jobs that survive crashes, money that can't be counted twice, and code that doesn't wake anyone up at 3 AM.

I work best owning a product end to end — schema to deploy, billing to monitoring — and writing down what I learn along the way.

Languages
TypeScript, SQL
Backend
Node.js, Express, oRPC, Zod, Drizzle, Prisma
Data & queues
PostgreSQL, Redis / Valkey, BullMQ, pgvector
AI
OpenRouter, Gemini, OpenAI, LangChain
Apps
React, Next.js, Astro, React Native, Expo
Infra
AWS, Cloudflare, Docker, Sentry, PostHog
05 — Contact

Building something ambitious in AI or SaaS? I'd like to hear about it.

tpbh230@gmail.com