Compiling systems…
BALRAM SINGH ARCHITECT
01 / The Person
PORTFOLIO
7+ years architecting fintech, logistics & AI systems. See the work behind the offer.
Enter
★ Start here
02 / The Offer
TECH
PARTNERSHIP
Your technical co-founder — zero cost to build. Equity vested on milestones. 2–3 partnerships a year.
Enter
BALRAM SINGH
[ SOLUTIONS ARCHITECT · AGENTIC AI & CLOUD PLATFORMS ]

I ARCHITECT AI-NATIVE SYSTEMS THAT SCALE.

Over 7 years designing and delivering enterprise-grade platforms across fintech, logistics and AI — backend engineering, cloud infrastructure (AWS / Azure / GCP) and multi-agent AI systems built on LangGraph and MCP, in production, not in demos. Currently Solutions Architect at TanServ, New Delhi.

Download Resume
Balram Singh — Solutions Architect & AI Engineer, New Delhi
0
Years Experience
0
Daily API Requests
0
Uptime Delivered
0
Projects Delivered
Multi-Agent Systems·MCP·LangGraph·AWS·Azure·GCP·Kubernetes·Java 17·Spring Boot·Python·LLM Evals·Kafka·System Design· Multi-Agent Systems·MCP·LangGraph·AWS·Azure·GCP·Kubernetes·Java 17·Spring Boot·Python·LLM Evals·Kafka·System Design·
Career Journey

SEVEN YEARS,
ONE THROUGH-LINE.

2025
● Current
TANSERV
Solutions Architect / Tech Manager
Architected microservices serving 100K+ daily requests at 99.95% uptime; built predictive-AR & multi-agent AI workflows (LangGraph, MCP) cutting manual ops 60%; migrated monolith → microservices for −75% deploy time.
2023
SAFEXPRESS
Senior Software Engineer
Built APIs handling 50K+ daily transactions; optimized Oracle queries 800ms → 120ms; real-time notifications via WebSockets + Redis; test coverage 40% → 85%.
2023
MASTERCARD
Full Stack Developer (Contract)
Enterprise solutions for global payment processing; shipped 12+ features across 3 sprints with zero production incidents; −45% dashboard load time.
2021
TECHGINIA GLOBAL
Java Developer / Scrum Master
Delivered 8 API modules serving 20K+ daily users at 99.7% uptime; deploy time 2h → 15min; mentored 2 developers to promotion.
2020
CONCENTRIX
Software Engineer
Delivered 150+ API endpoints across multiple projects; 98% code-review approval; contributed to 3 major web-application releases.
Selected Work

SYSTEMS IN
PRODUCTION.

Open the architecture — how Inebura is actually built
The problem

AR teams work overdue lists, not risk: collections are reactive, reconciliation is manual, and every enterprise client demands hard data isolation and audit-grade compliance. Inebura had to be a multi-tenant SaaS that predicts late payment before it happens — and prove its security posture (ISO 27001 / SOC 2 aligned) while doing it.

The system — 100% Google Cloud, zero-downtime
Edge
Cloud Load Balancing · Google-managed TLS · Cloud Armor — tenant resolved per request (subdomain → schema), JWT carries the tenant claim so a token can never cross tenants.
Application
React 19 SPA (three role-gated apps in one code-split build) · two Spring Boot / Java 21 services on virtual threads — AR, invoicing, comms, BI on one; identity, ledger & tenancy on the other. Containers on GKE — autoscaled, zero-downtime rolling updates.
Data
Cloud SQL for MySQL — regional HA (synchronous standby, automatic failover) + read replica · schema-per-tenant with per-client encryption keys · Memorystore for Redis + Caffeine two-level cache · Managed Service for Apache Kafka event streams driving real-time dashboard engines.
Integrations
Cloud Storage signed URLs · Exchange & SMTP per client · scheduled SFTP document ingestion · WebSockets with token + tenant binding.
Decisions that mattered
  • Schema-per-tenant over row-level flags. Each tenant gets its own MySQL schema and its own encryption key — isolation you can prove to an auditor, not just assert in a WHERE clause. Per-tenant migrations run app-managed.
  • Kafka instead of polling. Dashboard metrics are computed by event-driven engines on the stream — dashboards stay real-time without hammering the OLTP database.
  • Failover as a design, not a hope. Cloud SQL regional HA: primary down → the synchronous standby is promoted automatically, the read replica keeps reads flowing, and the recovered node rejoins the chain. Rehearsed, automated, documented in the DR plan.
  • Audit as architecture. Entity versioning across ~150 entities maps directly onto ISO 27001 Annex A / SOC 2 controls — compliance evidence is generated by the system, not assembled after the fact.
The AI stack — three layers, explainable by design
Layer 01 · Identity
BEHAVIOUR DNA

A continuously-updated behavioural profile per customer — payment pattern, communication preference, dispute profile, risk trend — computed deterministically from signals already in the platform.

Layer 02 · Prediction
DELAY PREDICTION

Per-invoice delay probability (0–100) with the top-3 contributing reasons and a named action, owner and deadline. Every score explains itself — no black boxes in a finance workflow.

Layer 03 · Orchestration
COLLECTIONS PLAYBOOK

An autonomous engine that builds a per-customer, multi-step contact & escalation plan — who to contact, when, on which channel — consuming the two layers below it.

Logistics Middleware · Active

INTEGRABLY

One API for every carrier — payload transformation, protocol bridging (REST, SOAP, SFTP, webhooks, CSV, EDI), smart routing and unified tracking.

1API · all carriers
24/7observability
Platform · Completed

COMIXFLIP

Digital comics platform with a custom reader, AI recommendations and offline PWA support.

10K+users
50K+comics
Infrastructure · Active

API GATEWAY

Enterprise gateway for auth, rate limiting, routing and monitoring across microservices.

100K+req/s
99.99%uptime
Agentic AI · Personal

TASK ORCHESTRATOR

Agentic system that parses natural-language tasks and autonomously orchestrates execution.

24/7autonomous
0babysitting
Capabilities

THE FULL
STACK.

01

Cloud & Infra

  • AWS
  • Azure
  • GCP
  • Kubernetes
  • Docker
  • Terraform
  • CI/CD
02

Backend

  • Java 17
  • Spring Boot
  • Python
  • Node.js
  • REST
  • GraphQL
  • Microservices
03

AI & Agentic

  • Multi-Agent Orchestration
  • MCP
  • LangGraph
  • LangChain
  • RAG
  • LLM Evals & Observability
  • OpenAI / Claude APIs
  • Vector DBs
04

Data & Messaging

  • PostgreSQL
  • Oracle
  • Redis
  • Kafka
  • WebSockets
  • EDI
05

Architecture

  • System Design
  • Security
  • Scalability
  • Event-Driven
  • API Design
06

Practices

  • Scrum
  • DevOps
  • Mentoring
  • Tech Leadership
  • Roadmapping
Contact

LET'S BUILD
SOMETHING SCALABLE.

Emailbalram@integrably.com LinkedIn/in/balramsingh-tech
LocationNew Delhi, India
FoundersAsk about the free Tech Partnership
[ TECH PARTNERSHIP — 2–3 SLOTS A YEAR · NOW OPEN ]

YOU DREAM IT. I BUILD IT.

I take on 2–3 startups a year as technical co-founder. If yours is one of them, I own the complete technical side — architecture, engineering, cloud & AI — at zero upfront cost. In return: 10–25% equity, agreed in writing before I build, vested only as milestones ship.

0
Years Architecting
0
Daily API Requests
$0
To Start Building
The Deal

THE DEAL

// no invoices. no retainer. terms in writing before a line of code.
You bring

The idea, the vision and the domain expertise. You know the problem cold.

I bring

The complete technical build — architecture, engineering, cloud & AI. End to end.

Cost today

$0

Nothing upfront. No hourly. No retainer. Ever.

Equity

10–25%

Agreed upfront, vested as milestones ship — not on promises.

How it works

HOW IT WORKS

01

Apply

Tell me the idea & the problem it solves.

02

Scope

A short call to align on vision & fit.

03

Build

I own & ship the full tech stack with you.

04

Grow

When it succeeds, we talk equity as partners.

The Build Flow

FROM IDEA TO
LIVE PRODUCT.

// the full solution flow — prototype to scaled, supported product
01
Prototype

Prototype & Website

We start with something real — a working prototype and a landing site. See it, click it, validate the idea fast before any heavy investment.

02
Iterate

Recursive Build & QA

Build → test → refine, on a loop. Every module goes through recursive software checks until it's solid, secure and genuinely production-grade.

03
Stack

Tech & Cloud Selection

I pick the right technologies and cloud — AWS, Azure or GCP — matched to your product's scale, compliance and budget. Deliberate choices, not hype.

04
Design

Full Product Architecture

The complete end-to-end architecture — services, data flow, APIs, security and scaling paths — mapped out before a line of production code is written.

05
Intelligence

AI Implementation

Agentic AI, automation and LLM workflows woven in exactly where they create leverage — smarter operations and real outcomes, not AI for its own sake.

06
Efficiency

Cost Impact & Optimization

Continuous cost analysis across cloud and every service — right-sizing, autoscaling and cutting waste, so the product performs without ever overpaying.

07
Ship

Delivery & Launch

Ship to production with CI/CD, monitoring and observability in place. A clean, reliable go-live — never a hand-off-and-vanish.

08
Care

Support & Managed Service

After delivery I stay on — service management, ongoing support, updates and iteration as you scale. The product keeps getting better over time.

Who it's for

WHO IT'S FOR

// 2–3 partnerships a year. a real idea and the drive to see it through.

SOLO FOUNDERS

You have the vision and the domain expertise — but no technical co-founder or team.

EARLY STARTUPS

Pre-seed teams who need serious engineering muscle before they can raise or hire.

ENTREPRENEURS

Idea people with a real problem worth solving and the commitment to build it right.

Straight answers

THE QUESTIONS
YOU'RE THINKING.

Who owns the code and the IP?

Your company. 100%, from day one. Everything I build — code, architecture, infrastructure — belongs to the startup, not to me. If we ever part ways, it all stays with you.

How much equity, exactly?

10–25%, depending on stage and scope. It's agreed in writing before I write a line of code, and it vests against shipped milestones — a working prototype, a live product, real users — not against time or promises.

What happens if the startup fails?

You owe nothing. There are no invoices to settle and no clawbacks — unvested equity simply lapses, the code stays yours, and we part on good terms. The downside risk is mine by design.

What's the catch?

Selectivity. I take 2–3 partnerships a year and say no to most applications — I only build what I genuinely believe can win, because my only payday is your success.

How is this different from a dev agency?

An agency bills hours and hands off. I'm a partner: architecture, engineering, AI, cloud, launch and iteration — aligned on the outcome, not the invoice. I stay on as the product scales, because my equity vests when it does.

Apply

PITCH ME
YOUR IDEA

Takes about 2 minutes. The more specific you are about the problem and where you're at, the better. If there's a fit, we start scoping this week.

  • ✓ Zero cost — no invoices, ever
  • ✓ You own 100% of code & IP
  • ✓ 10–25% equity, milestone-vested
Balram Singh — your technical co-founder
Balram Singh · your technical co-founder
Sends straight to my inbox — I read every application.

APPLICATION IN.

Thanks — your pitch is with me. If it's a fit, you'll hear back within a few days to book a discovery call.