Products people file with
Maintaining and modernizing commercial tax software — H&R Block Professional and ATX — for the professionals and firms whose whole season runs on it.
I'm Monish Bista — a senior software engineer in Marietta, Georgia. For ten years I've built and maintained enterprise software across tax and healthcare technology, most recently the commercial tax products that accounting firms depend on every filing season.
10+ years
enterprise software engineering
C# / .NET
and SQL Server, daily
Tax & health
two regulated domains
Agentic AI
in the engineering workflow
My work sits where the rules are written by someone else and the deadline isn't negotiable — tax code that changes every year, healthcare data that has to move correctly the first time. A failed nightly job isn't an alert; it's somebody's return or somebody's claim.
Maintaining and modernizing commercial tax software — H&R Block Professional and ATX — for the professionals and firms whose whole season runs on it.
ETL pipelines that used to be spreadsheets, deployments that used to be checklists, reports that used to be somebody's Thursday afternoon.
Agentic AI applied to real engineering work — code analysis, documentation, defect triage, release readiness — inside secure SDLC and compliance boundaries.
Commercial tax products — H&R Block Professional and ATX — built and maintained end to end, plus the AI-augmented workflows I introduced across several tax software teams.
Healthcare applications in C#, .NET, and SQL Server, with SSIS pipelines moving high volumes of claims data and automation replacing manual operations work.
EDI testing with trading partners, X12 and HCFA 1500 mapping, and the first scripts I wrote to kill repetitive work.
I studied computer science at Kennesaw State and went straight into healthcare software, starting as an intern mapping X12 claim files and staying because the problems kept being interesting. Claims data is messy, the rules change, and every integration has a partner on the other end with their own idea of what a valid file looks like. Tax software turned out to be the same shape of problem on a harder calendar.
Lately much of my attention has gone to putting agentic AI where it actually helps — summarizing dense compliance rules, generating documentation nobody has time to write, triaging defects faster — without loosening the review and security standards that make the work trustworthy.
I work best on teams where engineers review each other honestly and documentation counts as part of the deliverable. I'd rather spend an afternoon making a deployment boring than spend a night fixing an exciting one.
Email is the fastest way to reach me. I'm happy to talk about backend work, tax and healthcare systems, or putting AI to work inside a secure SDLC.