ProjectsWritingAboutContact

About

Full-stack engineer. Ships and runs production systems alone.

I build full-stack production systems and run them. The current one is leohydra.com, a Dubai art studio I run from Mumbai. We do 20 to 30 orders a day, and every limited edition print drop sells out inside a week of going live. The whole codebase is mine. So is the on-chain payment verifier, the admin panel, the transactional email pipeline, the Meta Pixel and CAPI dedup, and the security audit migrations. I run the ads too, manage the printers, and ship the parcels.

B.E. in Electronics & Telecommunication from Smt. Kashibai Navale College of Engineering, Pune. That was the full time campus degree, 2016 to 2020. After that I picked up the IIT Madras Online Diploma in Data Science, 2021 to 2022. Five roles between 2023 and now: a Canopy Singapore data ops contract staffed via Recro, a Cointab data analyst job, two internships, and the current Leo Hydra engagement. I've been engineering and running ops on the LeoHydra stack since late 2025.

I'm strongest at production e-commerce engineering, growth instrumentation (server side Pixel and CAPI, dedup aware), security hardening, and running ops alone. I'm comfortable being the second or third engineer at a startup that has no infrastructure team yet.

Experience

Five roles between 2023 and now.

Oct 2025 to Present

Remote (Mumbai / Dubai)

Freelance

Freelance Engineer + Operator

Leo Hydra Studio

  • ·Sole operating team for a one-principal art business based in Dubai. Hand-coded the platform from scratch and run everything else from Mumbai. No template, no inherited codebase.
  • ·Current scale: 20 to 30 orders per day. Every limited-edition print drop sells out within 7 days of going live. Repeat-buyer rate among collectors who own at least one print.
  • ·Architected and wrote the production e-commerce platform at leohydra.com from scratch. Stack: Next.js 14, Supabase Postgres, USDT-on-Polygon checkout with on-chain payment verification, bank transfer with admin manual confirm, ~12 tables, ~30 migrations, ~22 API routes, ~50 React components.
  • ·Server-side Meta Pixel + Conversions API across every conversion surface (checkout / forms / newsletter / product pages / PageView), dedup-aware via shared eventID. Transactional email pipeline (Resend, 6 lifecycle kinds, audit-logged with admin resend UI).
  • ·Three documented security audit rounds + 5 new migrations (027 to 031): DB-backed rate limiter, admin brute-force lockout, CSRF Origin checks, PII redaction, HMAC-secret isolation, CSP + security headers.
  • ·Operations: run Meta ad campaigns, manage limited-edition print production with printing houses, coordinate Dubai-warehouse fulfillment, hire and direct video editors, sound design + multi-platform posting.

Nov 2024 to Jan 2025

Singapore (remote, via Recro)

Full-time contractor

Data Operations Analyst

Canopy

  • ·Full-time on Canopy's data ops team (private-wealth SaaS, Singapore), staffed through Recro India as the legal employer.
  • ·Built automated SQL + Python pipelines for ~3,000 daily/monthly financial transactions across the portfolio system, ensuring 100% reconciliation.
  • ·Automated data validation and error-fixing in Python; raised report accuracy ~15%.
  • ·Resolved portfolio discrepancies during corporate-action events with 100% accuracy.
  • ·Partnered with business teams on transactions-per-user, retention anomalies, and portfolio-performance KPIs.
  • ·Role concluded due to contract ramp-down.

May 2024 to Aug 2024

Mumbai, India

Full-time

Data Analyst

Cointab

  • ·Built automated multi-stage financial-reconciliation workflows for client Gameskraft's fraud-detection pipeline; saved ~100 analyst-hours/month.
  • ·Configured in-house tooling using Python and SQL to automate reconciliation logic for financial audits.
  • ·Analyzed customer transaction flows to detect anomalies, improving fraud-detection accuracy.
  • ·Designed BI reports on churn cohorts and transaction-volume trends with marketing and sales.
  • ·Role ended in a company-wide layoff (Aug 2024).

May 2023 to Mar 2024

Remote, India

Internship

Data Science Intern

Happymonk AI Labs

  • ·Built CNN-based object-detection models (YOLO, Faster R-CNN) reaching 98% inference accuracy by aggregating multi-source data.
  • ·Led large-scale image annotation (100,000+ images) with rigorous QA pipelines.
  • ·Deployed real-time detection systems on 50+ concurrent video streams achieving <200 ms latency with 95%+ inference accuracy.
  • ·The data-science department was shut down before conversion to full-time.

Jun 2023 to Sep 2023

Remote, India

Internship

Research Intern

Healthcare Technology Innovation Center

  • ·Contributed to ML/AI techniques for vascular health assessment using ARTSENS (image-free ultrasound system).
  • ·Developed Python scripts to automate ultrasound data screening and generate motion-mode images.
  • ·Designed and evaluated ML models for carotid-artery wall dynamics.
  • ·Four-month internship, completed on schedule.

Skills

Languages

TypeScriptJavaScriptPythonSQLHTML / CSS

Frameworks

Next.js 14 (App Router)ReactTailwind CSSFastAPINode.jsPandasNumPyscikit-learnPyTorchTensorFlowXGBoost

Tools & Infra

VercelSupabase (Postgres + RLS)DockerGitHub ActionsPolygon RPCethers.jsPower BITableauGit

ML / AI

RAG / LangChain-style pipelinesLLM-native workflow designCNN (YOLO, Faster R-CNN)Object detectionRandom ForestXGBoost

Specializations

Production e-commerce engineeringMeta Pixel + CAPI (server-side, dedup-aware)Meta ad campaign operationsWeb3 payments (USDT on Polygon)Security hardening (rate limiting, brute-force lockout, CSRF, CSP, HMAC isolation)Solo engineering ownership end-to-endVendor & supply-chain management

Education

2021 to 2022

CGPA: 8.2

Diploma in Data Science (Foundational + Diploma levels of the IIT Madras Online Degree program)

Indian Institute of Technology (IIT), Madras

2016 to 2020

CGPA: 7.0

B.E. in Electronics & Telecommunication

Smt. Kashibai Navale College of Engineering, Pune