I ship and automate production systems end to end. Payments, AI products, data, infra, usually alone, with agents doing much of the engineering.

SuyogParkhi

ScrollPayments · AI · Data · InfraAvailable 2026
(01) About

I build consumer apps and the systems under them: Saarthi 1M+ downloads and Soundpay 50K+ downloads at Sozo Labs, and payment‑data infrastructure at Juspay for rails that carry half of India’s UPI traffic 10,000 TPS.

One engineer, every layer

Founding engineer at Sozo Labs, previously Juspay. Backend, infra, data, ML and mobile. Most of the engineering runs through agent harnesses, MCP servers over production and AI review on every diff, which is how one person covers a whole stack.

Bengaluru
IN
1M+
01 · Saarthi

Downloads on Google Play. An AI companion in 10+ Indian languages.

50K+
02 · Soundpay

Downloads on Google Play. UPI payment alerts, read aloud.

10K
03 · Peak, UPI

Transactions per second at the PSP where I built payment‑data infra.

50%
04 · UPI share

Of India’s UPI traffic runs on the rails that infra serves.

(02) Selected work

Apps andsystems(03)

01SaarthiSozo Labs · founding engineerLive · Google Play
Fig. 01 · 10+ Indian languagesLive

Saarthi

AI companion · Android

An AI astrology and spiritual companion. People talk to it by chat and voice in more than ten Indian languages. I built and run its backend, payments, data stack and infrastructure, and moved it from DigitalOcean to AWS with live replication and no data loss.

Google Play ↗
Downloads
1M+
Languages
10+
Modes
chat + voice
Cloud
DigitalOcean → AWS
FlutterFastAPIGeminiLiveKit voiceDocumentDBClickHouseGrowthBookAWS ECSTerraform
02Juspay · UPI DataProduct engineerDec 2025 — Jun 2026
0 txnsFig. 02 · 10,000 / s

Juspay
UPI Data

Product engineer

Payment‑data infrastructure at a PSP processing about 50% of India’s UPI traffic, inside NPCI‑regulated flows. I designed UPI‑Data‑OCD, a Nix and Terranix framework that turns ~1,000 lines of per‑tenant Terraform into 30–80 lines of typed Nix, and used it to stand up analytics for Kotak Bank and SuperMoney.

Terraform per tenant
1,000 → 30–80 lines
Tenant onboarding
days → 1 command
Long‑lived AWS keys
removed
HaskellKafkaCassandraClickHouseNix
03BingeClip AISDE internJan — May 2025
Fig. 03 · queue depth → workersKEDA

BingeClip
AI

SDE intern

The backend for a WhatsApp bot that generates AI video. Every user gets an isolated Redis queue, every client its own RabbitMQ workers, and KEDA scales the workers on queue depth, so one heavy user never stalls another.

Queues
isolated per user
Workers
per client
Scaling signal
queue depth
RedisRabbitMQKEDAKubernetes
(03) Inside Sozo Labs
Internal tools and current work

Alsobuilding

  1. 01

    AI ad tracker

    Meta Ad Library · LLMLive

    Tracks competitor ads from the Meta Ad Library as they start, stop and restart. It transcribes and tags each ad, analyses the set and drafts an ad strategy from it.

  2. 02

    AI video pipeline

    Script → video adLive

    Turns a script and a character into a finished video ad for Reels and Shorts, in Hindi and other Indian languages. The team makes ads with it in batches.

  3. 03

    3D AI characters

    Unity · realtime voiceBuilding

    An animated 3D character you talk to and interact with. It runs in Unity inside a native iOS app, with realtime voice. Not released yet.

  4. 04

    Tax filing agent

    GST · income taxBuilding

    An agent that prepares and files Goods and Services Tax and income tax returns for Indian taxpayers.

(04) ResearchNotes and side work

Research
first

Before I build something new, I research it and write it up: dated sources, real numbers, and a clear line between evidence and vendor claims. On the side I built find, a private daily feed that replaces algorithmic social media for me. It reads sources I trust, ranks each item against a written profile of what I care about, and every edition ends.

  1. 01 · ResearchAgent fleets that ship code
  2. 02 · ResearchProduction errors to fixes
  3. 03 · ResearchAI pipelines for 3D characters
  4. 04 · StudyAgentic AI systems
  5. 05 · Side projectfind, a private feed
(05) MethodHow I work with agents
A

Parallel research harnesses with specialised subagents and custom Claude Code skills, run as daily practice.

JavaPythonRustHaskellTypeScriptC++KotlinSQL(Languages)
KafkaClickHouseCassandraPostgresRedisRabbitMQMongoDB(Data)
AWSKubernetesNixTerraformGeminiClaudeUnityLiveKit(Infra · AI)
(06) RecordWork and study

2026

  • Sozo Labs

    Jul 2026 — Now

    Founding engineer, all technology · Bengaluru

  • Juspay

    Dec 2025 — Jun 2026

    Product engineer, UPI Data · Bengaluru

2025

  • Sozo Labs

    Jul — Dec 2025

    First engineer: payments, AI companion, DigitalOcean to AWS · Remote

  • BingeClip AI

    Jan — May 2025

    Software development engineer, intern · Remote

2022

  • RCOEM

    2022 — 2026

    B.E. Computer Science, Shri Ramdeobaba College of Engineering and Management · Nagpur

  • IIT Madras

    2022 — 2025

    Diplomas in Programming and Data Science (BS Data Science) · Chennai

(07) ContactFounding and early engineering roles

Let’sbuild

© 2026 Suyog ParkhiBengaluru --:-- ISTBack to top ↑