About the proprietor
I'm an Associate Software Engineer at Finzly, working on payment processing and cloud infrastructure for multi-tenant banking platforms that serve 50+ financial institutions. Day to day that is Java, Spring Boot and AWS, and a fair amount of talking to product managers and banking clients about what they actually need.
Outside that I build small things that do one job well — voice agents that answer the phone, WhatsApp bots that take orders, tools that run on free tiers. Most of it started on a Saturday.
The works
Everything below shipped, or at least ran once on a Saturday.
-
No. 002
AI Call Assistant
shippedA self-hosted AI phone assistant. Someone calls your number, an AI answers, holds a real conversation, books appointments and takes messages.
Read the full account -
No. 001
SaaS Opportunity Scan
weekend buildA weekend research pass on where the actual B2B pain is — and proof that X trends are the wrong instrument for finding it.
Read the full account
Particulars & résumé
Experience
-
Associate Software Engineer
Finzly — Chennai, India Apr 2026 — Present
- Leading development of payment processing features and cloud infrastructure for multi-tenant banking clients serving 50+ financial institutions.
- Work directly with product managers and banking clients to gather requirements, clarify business workflows, and translate them into scalable technical solutions.
-
Software Engineering Trainee
Finzly — Chennai, India Aug 2024 — Mar 2026
- Cut Payment Blotter API response time from 2–3 seconds to 300–600 ms through query optimization.
- Built a scheduled Holiday Job that ingests third-party holiday data, stores files in environment-specific S3 buckets, and updates multi-tenant databases.
- Built an HTML editor for client notification templates with automatic synchronization to S3 and GitHub.
- Replaced Javers with audit logging via SQL triggers and AWS Lambda, reducing container overhead and enabling frontend audit visibility.
- Integrated the Twilio Lookup API in Java to validate global phone numbers, and extended Angular UI components for in-app navigation.
-
Intern
Samsung PRISM — Virtual Jan 2023 — Aug 2023
- Worked on unsupervised synthetic-to-real domain adaptation for image segmentation.
- Investigated existing segmentation solutions and datasets including SYNTHIA, Cityscapes, ADE20K and COCO.
- Explored visualization techniques to make domain adaptation results more interpretable.
Education
-
B.Tech, Computer Science
SRM Institute of Science and Technology — Chennai, TN Aug 2020 — May 2024
Correspondence
- E-mailgumanpratapsinghparmar@gmail.com
- LinkedInlinkedin.com/in/gpsinghh
- GitHubgithub.com/Gumanpratapsingh