Sagar Rathod, systems architect at Virus HQ

Architecture that scales, costs less, and stays up.

Sagar Rathod · Solutions & Systems Architect · Virus HQ: 9+ years designing and shipping fintech-grade backend, cloud & data platforms

Virus HQ is my independent architecture and engineering practice. I work with companies, startups, and founders from Australia and Japan to the US and Mexico, from first PoC to platforms serving 100K+ users. I design the target architecture, then ship the code, run the migration, and own the result. For larger builds, I bring in a small bench of trusted senior engineers.

Available for: solution architecture & builds · fractional architect retainers · cloud cost reviews · Staff / Principal roles (remote)

View My Work

Fintech · Trading · SaaS · EdTech · E-commerce · Shipping production systems since 2017

About & availability

What I do for clients, how I work with US/EU teams, and when I'm available.

Virus HQ is a solo-led engineering practice for teams where infrastructure directly drives revenue, margin, and trust. I'm Sagar, a solutions and systems architect with 9+ years in backend, platform, and data engineering, mostly in fintech and trading. I've migrated a ~1,200-endpoint .NET monolith to Go services, rebuilt trading and education platforms end to end (backend, frontend, DevOps and infrastructure), cut cloud bills by more than half, and built AI pipelines that review thousands of calls a day. Work with me directly for architecture and hands-on delivery, or engage Virus HQ as a small team when the scope needs more hands. Either way, I stay accountable for the outcome.

  • ✓Hands-on ownership: I ship production code, run the migration, and own the result. No slide decks.
  • ✓Cost-aware by default: architecture that cuts cloud spend without trading away performance or reliability.
  • ✓Security built in: hierarchical RBAC, PKI signing, IDOR protection, and proper secrets management as defaults, not add-ons.
  • ✓Works in your time zone: remote since 2017 with clients from Australia and Japan to the US and Mexico, plus clear written updates between calls.
  • ✓Ahead-of-schedule delivery: every Virus HQ engagement has shipped ahead of the agreed timeline.
  • ✓Scales with your scope: solo for focused work, plus a vetted bench of senior engineers for larger builds.

Architecture & solutions engineering

I design the system, then build it with you. Clear scope, milestones, and outcomes so you know what you're buying before we start.

Core: architecture & solutions

Solution Architecture & End-to-End Build

From blank page to production: target architecture, technical decisions, then the backend, frontend, infrastructure, CI/CD, and observability, delivered solo or with a small Virus HQ team.

Typical deliverables

  • • Architecture blueprints & technical decisions record
  • • Production-ready backend & APIs
  • • Scalable frontend
  • • Automated CI/CD + observability stack
💰 From $15,000 · paid per milestone ⏱ Typical timeline: 1.5-12 months

Re-Architecture & Platform Rebuilds

Redesign and rebuild core systems for platforms that have outgrown their architecture, delivering scale, reliability, and cost efficiency.

Typical deliverables

  • • Complete architecture redesign
  • • Backend & infrastructure implementation
  • • Real-time systems (WebSockets, event-driven)
  • • Zero-downtime cutover & stabilization
💰 From $12,000 · project-based ⏱ Typical timeline: 4-16 weeks

Fractional Staff / Solutions Architect

Staff-level architectural leadership without full-time headcount: design reviews, technical direction, team mentorship, and accountability for system health.

Typical deliverables

  • • System design blueprints & reviews
  • • Team mentorship & engineering standards
  • • Scalability, reliability & cost roadmaps
💰 $5,000-$12,000 / month retainer ⏱ Typical timeline: Ongoing · 3-month minimum

Also available

Cloud Cost Recovery

Stop cloud bleed. Deep audit, architectural redesign, and hands-on implementation that cuts AWS/GCP spend by 30-60% through better design, not just discounts.

Typical deliverables

  • • Preliminary cost scan with savings estimate
  • • Hands-on cost-saving implementation
  • • Long-term scaling & sustainability roadmap
💰 Preliminary scan (8-12 hrs), then 25-35% of first-year savings ⏱ Typical timeline: 3-10 weeks

Legacy Modernization & Performance Fixes

Incremental modernization of brittle or monolithic systems, plus focused fixes for latency, throughput, concurrency, and reliability.

Typical deliverables

  • • Target-state architecture & phased migration plan
  • • Database/query tuning, caching & queuing improvements
  • • CI/CD pipelines with zero-downtime releases
💰 From $6,000 · project-based ⏱ Typical timeline: 2-10 weeks

AI Pipelines & Workflow Automation

Production LLM and speech pipelines that remove manual review work: transcription, classification, validation, and confidence-based routing to humans.

Typical deliverables

  • • Use-case & data audit
  • • ASR / LLM pipeline on AWS Bedrock or equivalent
  • • Confidence thresholds & human-in-the-loop review
  • • Accuracy, cost & latency monitoring
💰 From $8,000 · project-based ⏱ Typical timeline: 3-8 weeks

Data Pipeline Optimization

Turn slow, brittle batch jobs into fast, reliable, observable distributed pipelines.

Typical deliverables

  • • Pipeline audit & bottleneck analysis
  • • Re-architected processing engine
  • • Multi-source ingestion & orchestration
  • • Monitoring, alerting & runbook
💰 From $5,000 · project-based ⏱ Typical timeline: 2-5 weeks

Cloud cost review

A low-risk way to start: find the waste first, pay out of the savings.

Not sure where to start? Begin with a cloud cost review.

A low-risk first step: a preliminary scan of your cloud setup (typically 8-12 hours of work) that shows where the money goes and what can be cut. If you go ahead, implementation is priced as a share of what you save, so the work pays for itself.

  • 1. Preliminary scan: billed hourly, typically 8-12 hours
  • 2. Written findings with a savings estimate
  • 3. Implementation at 25-35% of verified first-year savings

Estimate your savings

$
$6,000-11,600
Potential monthly savings
$72,000-139,200
Potential first-year savings
$18,000-48,720
Implementation fee (25-35% of savings)
$46,800-104,400
You keep in year one

Based on the 30-58% reductions I've delivered. Your real number comes from the preliminary scan.

How I work

A simple, transparent process from first call to hand-off.

01

Discover

A free 30-minute call to understand your goals, constraints, and current architecture. For larger scopes, a short paid architecture or cost review with written findings.

02

Design

A written proposal: target architecture, trade-offs, milestones, risks, and the payment % tied to each milestone. You know exactly what you're buying before work starts.

03

Build

Hands-on delivery in milestones with weekly written updates and demos. Production changes roll out safely, with zero-downtime cutovers wherever possible.

04

Hand-off

Code ownership transfers milestone by milestone as each one is paid. At the end you get every account and admin role, documentation, runbooks, and dashboards, and I remove my own access.

Projects

Deep dives with challenge, solution, and outcomes.

Monolith to Go: ~1,200-Endpoint Platform Migration

Deploys from 6h+ to under 30 min at 99.997% uptime

14 months

Large-Scale Fintech Platform

  • Challenge: A legacy .NET monolith with ~1,200 endpoints made releases slow, risky, and hard to scale for a partner platform with 100K+ active users.
  • Solution: Led the migration to Go + PostgreSQL services on AWS, with a dedicated performance-test environment (read replica + custom load generator at 4× peak) to de-risk the cutover.
  • • ~1,144 endpoints now running across 3 Go services
  • • Deploy time cut from 6h+ to under 30 minutes
  • • ~99.997% uptime (~15 min downtime/year)

Tech: Go • PostgreSQL • AWS • Terraform • GitHub Actions

Ground-Up Platform Rebuild: 50k+ Users, Sub-100ms Latency

Rebuilt in 9 months for 50k+ concurrent users

9 months

US-Based Trading & Education Platform

  • Challenge: A legacy patchwork codebase and infrastructure could no longer support scale or real-time requirements.
  • Solution: Sole ownership of the complete rebuild, infrastructure included: new backend (Node.js + Python + PostgreSQL), React frontend, real-time WebSocket infrastructure, CI/CD, and monitoring, delivered in 9 months.
  • • Full platform and infrastructure rebuilt from zero to production
  • • Sub-100ms WebSocket latency for live market data
  • • 50,000+ concurrent users at launch
  • • 14k+ requests/sec sustained
  • • Zero-downtime legacy cutover
  • • 99.997% uptime through peak volatility

Tech: Python • Node.js • PostgreSQL • React • WebSockets • AWS • Kubernetes • Terraform

Multi-Region Fintech: 58% AWS Cost Reduction

~$12-14K saved every month (58%)

3 months · 4 phases

Large-Scale Fintech Platform

  • Challenge: Cloud costs across multiple AWS regions were eroding margins, and the team lacked visibility into the root causes.
  • Solution: Full infrastructure audit followed by platform-wide architectural changes across Redis, RDS, S3, Glue, ALB, and ECS. Permanent structural fixes, not one-off discounts.
  • • ~58% reduction in monthly AWS spend from a $20-25K/month baseline
  • • 75% reduction in Redis costs
  • • 64% reduction in RDS costs
  • • 86% reduction in S3 spend
  • • No performance degradation

Tech: AWS • Terraform • Redis • RDS • ECS • S3

Payout Pipeline: 12 Hours to 30 Minutes

Nightly payout run cut from 12 hours to 30 minutes

Large-Scale Fintech Platform · 2 phases · Phase 2 base in 3 weeks

  • Challenge: A 12-hour nightly payout and reconciliation batch built on stored procedures was failing to scale, creating operational risk and delays.
  • Solution: Two phases: first, optimized the existing MySQL stored procedures; then moved processing off the database, with data exported from MySQL to S3, processed in PySpark on AWS Glue, and written back to the database, migrating the legacy logic with zero downtime.
  • • Runtime reduced from 12 hours to 30 minutes (96% faster)
  • • Zero missed payouts during migration
  • • Full observability and alerting added
  • • 100% payout matching accuracy

Tech: PySpark • AWS Glue • MySQL • S3

AI Trade-Call Compliance Pipeline

60% less manual review across 8-12K calls a day

Large-Scale Fintech Platform · 5 months

  • Challenge: Thousands of recorded trade calls a day had to be validated against executed trades, a slow manual review process.
  • Solution: Built a speech-to-validation pipeline: Whisper plus an Indic-language ASR flow, Qwen on AWS Bedrock for call-to-trade validation, and confidence thresholds that route only uncertain records to humans.
  • • ~90% of 8-12K daily calls transcribed automatically
  • • ~55% of records auto-resolved
  • • Manual review effort reduced by 60%

Tech: Python • Whisper • AWS Bedrock • Qwen • LLM scoring

Trading Education for a New Generation: Live Classes at Scale

10× concurrency with 70% lower live-class latency

US-Based Trading Education Platform (sister brand) · 18 months

  • Challenge: A fast-growing sister platform with a younger audience and wider reach was hitting severe concurrency limits during live classes.
  • Solution: Re-engineered the real-time infrastructure with Node.js and Python: a horizontally scalable WebSocket layer, guaranteed message delivery, a robust quiz engine, and CDN optimization.
  • • 10× increase in concurrent users supported
  • • Quiz delivery failure rate reduced to near zero
  • • Live class latency reduced by 70%
  • • Content delivery 3× faster via CDN
  • • Zero outages during peak hours at significantly higher load

Tech: Node.js • Python • WebSockets • Redis Pub/Sub • AWS CloudFront • PostgreSQL

Enterprise Auth & RBAC: 650+ APIs Secured

650+ APIs and 100K+ users on one auth model

Large-Scale Fintech Platform · 1 month

  • Challenge: Inconsistent auth across services and a multi-level partner hierarchy created security risk and slowed down onboarding.
  • Solution: Hierarchical RBAC, surrogate login, JWT, PKI signing, and session policies, plus self-serve hierarchy management so partner firms manage their own employees. AWS WAF hardening and IDOR checks on every call.
  • • 650+ APIs secured
  • • 100K+ active users across 7 partner role types
  • • Service-to-service traffic for shared sub-services (Portfolio, Tagging, Ledger, Profile) used across multiple UIs moved onto Kong API Gateway
  • • Zero security incidents post-launch

Tech: Go • JWT • OAuth2 • AWS WAF • PKI • RBAC • Kong

Self-Serve Reporting & Experimentation Platforms

Apache Superset replaced for 4 teams

Large-Scale Fintech Platform · 1.5 months

  • Challenge: Reporting ran on Apache Superset with slow, fragile reports, and every feature rollout needed engineering involvement.
  • Solution: Built a schema-driven reporting platform (multi-source data, per-report timeouts, caching, async rendering) and a self-serve A/B testing and feature-rollout system for product and business teams.
  • • Replaced Apache Superset; used by 4 teams
  • • Product teams run rollouts and A/B tests without engineering tickets

Tech: Go • PostgreSQL • Redis • AWS

Fund House Data, Algo & Infrastructure Overhaul

CEO's manual data work automated end to end

US-Based Fund House · 4-year engagement (2019-2023)

  • Challenge: The CEO was spending excessive time on manual data tasks across interconnected self-hosted servers, with fragile infrastructure, unreliable alerting, and no robust backups.
  • Solution: Built systems to fetch and analyze market data, tooling to develop and test trading algorithms, signal-based triggers with real-time alerts, and on-site plus automated off-site backups. Managed and optimized the servers from 2019 to 2023.
  • • Dramatically reduced the CEO's manual workload
  • • Data and backtesting systems for algorithm development
  • • Reliable signal-based alerts for critical data events
  • • Stabilized multi-server self-hosted infrastructure
  • • Robust on-site + automated off-site backups

Tech: Python • Linux Servers • PostgreSQL • Docker • Terraform • Alerting Systems

Skills & Tech Stack

Stack depth plus proof points US/EU buyers look for.

Trust & impact

58%
Cloud Cost Reduction
96%
Pipeline Speed Improvement
50k+
Concurrent Users Scaled
650+
APIs Secured
<100ms
Real-Time Latency
99.997%
Uptime Delivered

At a glance

Focused on production systems where performance, cost, and reliability directly drive revenue, user experience, and competitive advantage.

Book a free 30-minute architecture call. No pitch, no slides: a clear diagnosis of your highest-leverage bottlenecks and a practical roadmap.

Remote since 2017 · Clients from Australia and Japan to the US and Mexico · Solo-led, team-backed when the scope needs it

Core Backend

GoPythonNode.js.NETC#

Cloud & Infrastructure

AWS (Expert)TerraformKubernetesDockerKong API GatewayServer Commissioning

Databases

PostgreSQLMySQLRedisRedshiftDynamoDBMongoDB

Frontend & Full-Stack

ReactTypeScriptTailwind CSS

Real-Time & Messaging

WebSocketsEvent-driven systemsSQS / SNSRedis Pub/Sub

Observability

GrafanaPrometheusCloudWatchPagerDuty

Security & DevOps

OAuth2JWTPKI SigningRBACGitHub ActionsWAFSnykSonarQube

Data & AI

AWS GluePySparkAirflowDatabricksAWS BedrockWhisper / ASRLLM pipelines

Long-Term Partnerships

Retainer-style work for teams that want steady ownership after a defined sprint or long-horizon modernization alongside your roadmap.

Platform & Product Engineering

Large-Scale Fintech Platform · 2022 - Present
Global Remote
Go, PostgreSQL, AWS, PySpark, Kong, Terraform

Long-term ownership of core re-architecture: moving a legacy .NET monolith to Go, PostgreSQL, and AWS services with clear boundaries, safer releases, and much faster delivery.

  • • Led migration of a ~1,200-endpoint .NET monolith to Go/PostgreSQL services
  • • Cut deploy time from 6h+ to under 30 minutes; ~99.997% uptime
  • • Built hierarchical RBAC & auth securing 650+ APIs and 100K+ users
  • • Built an AI trade-call compliance pipeline (60% less manual review)
  • • Re-architected payouts: stored-procedure tuning, then MySQL → S3 → Glue/PySpark (12h to 30m)
  • • Replaced Apache Superset with a schema-driven reporting platform
  • • Built self-serve A/B testing, loyalty, OTP, and notification services
  • • Led Kong API Gateway migrations and a 4×-peak load-testing environment
  • • Go-to reviewer for architecture changes; mentor to interns and peers

Data Automation, Algo Tooling & Infrastructure

US-Based Fund House · 2019 - 2023
Remote
Python, Linux Servers, PostgreSQL, Docker, Terraform, Alerting Systems

Long-term engagement helping a small, fast-moving fund eliminate manual data work, build algo research tooling, and stabilize critical self-hosted infrastructure.

  • • Built systems to fetch, store, and analyze market data
  • • Built tooling to develop and test trading algorithms
  • • Automated large volumes of the CEO's manual data tasks
  • • Built signal-based triggers and real-time alerting
  • • Managed, fixed, and optimized interconnected self-hosted servers
  • • Created on-site and automated off-site backups

Full-Stack Platform & Infrastructure Engineering

US-Based Trading & Education Platform · 2019 - 2022
Remote
Node.js, Python, React, AWS, Docker, Kubernetes

Started as the backend engineer and grew into owning the whole stack: frontend, DevOps, and systems administration across the platform and its sister brand.

  • • Rebuilt the entire platform and infrastructure from scratch
  • • Built sub-100ms WebSocket market data feeds for 50k+ concurrent users
  • • Scaled the sister brand's live classes 10× with zero peak-hour outages
  • • Implemented zero-downtime Docker/Kubernetes pipelines
  • • Owned ETL pipelines, cloud cost, and on-call rotations
  • • Created on-site and automated off-site backups

Infrastructure & Backend Engineering

Early-Stage SaaS & Fintech Startups · 2017 - 2019
Remote
Python, Node.js, AWS, PostgreSQL, Docker

Foundational work across multiple high-growth startups, building backend systems, cloud infrastructure, and data pipelines from scratch.

  • • Designed and shipped REST APIs & microservices
  • • Established production AWS infrastructure
  • • Built ETL pipelines for financial data feeds
  • • Introduced CI/CD and code standards

Writing

Notes from real engagements, anonymized.

FAQ

How engagements, payment, ownership, and access work.

Which time zones do you work with?

All of them. I've worked with clients from Australia and Japan to the US and Mexico, and I schedule calls and overlap around your core hours.

Can you share client names or references?

Most of my work is under NDA, so case studies are anonymized. I'm happy to go deeper on any of them on a call, within what those agreements allow.

Do you sign NDAs and contracts?

Yes. Every engagement starts with a contract that sets the scope, milestones, and the payment % for each milestone. NDAs are standard.

Who owns the code?

You do, once it's paid for. Work is signed off feature by feature, each milestone carries a set payment %, and when that payment clears the code for it transfers to you. Until then it stays with Virus HQ.

How does payment work?

I invoice per milestone and payment is due within 7 days. Wire transfers are fine.

How do you handle access to our systems?

Access runs through your SSO and user management. I start as admin or co-admin so I can move quickly, and on completion I transfer every account and role to your team and remove myself from the system.

How is a cloud cost review priced?

The preliminary scan is billed hourly and usually takes 8-12 hours. If you move ahead, implementation is priced at 25-35% of verified first-year savings, so you only pay out of what you save.

Is it just you, or a team?

I lead every engagement and do the hands-on architecture and engineering. For larger builds I bring in trusted senior engineers from my network, and I stay accountable for the result.

Contact

Send a message, email, or book a call, whichever is easiest for you.

Let's design the system your next stage of growth needs.

Book a free 30-minute architecture call. Describe what you're building or what's breaking, and you'll leave with a clear diagnosis and next steps, no strings attached.

hello@virushq.trade

© 2026 Sagar Rathod · Virus HQ · Solutions & Systems Architect
Planning a build or a re-architecture? Let's talk
Free 30-min architecture call · solution design · builds · cost reviews