Software Engineering Blog, page 3
Notes on software engineering from the Atyantik practice.
Latest posts
Kalpesh Patel
Can you trust AI-generated code? The checks that earn it
Can you trust AI-generated code? Match each way generated code fails to the review, test, scan or benchmark that catches it, plus a runnable CI gate.
Read the postHardip Patel
Should I build an MCP server for my platform, and what goes in it first?
Should I build an MCP server? Three tests to decide, the measured cost of one tool per endpoint, and which tools to release first and which to keep off.
Read the postHardip Patel
Build a B2B ecommerce AI assistant or buy one: three tests on your own portal
Should you build a B2B ecommerce AI assistant or buy one? Three tests on your portal decide: who releases a price, how fast it comes, how alike parts are.
Read the postAyush Makwana
MCP server security best practices, in the order a request meets them
MCP server security best practices in request order: what the spec requires, what it leaves to you, and the public failures each check would stop.
Read the postAyush Makwana
Glean alternatives: when building your own internal AI assistant beats a per-seat licence
Glean alternatives weighed from each vendor's own pages: Microsoft 365 Copilot, Gemini, Rovo, Slack enterprise search, and when a narrow build wins.
Read the postHardip Patel
Why RAG fails in production, and the service fixes that come before retrieval
Why RAG fails in production after a good demo: seven service defects that break it, how to spot each one, and the fixes to make before touching retrieval.
Read the postHardip Patel
Self-hosting Cloudflare Clef-flash with SGLang: context, GPU memory and cost
Self-hosting Cloudflare Clef-flash with SGLang: the launch command, the real context limits, GPU memory by arithmetic, and when a GPU beats Workers AI.
Read the postAyush Makwana
Microsoft-Decision-1 Foundry decision automation: the integration work, and who should do it
Microsoft-Decision-1 Foundry decision automation for leaders: what the integration involves, in-house or partner, what to ask, thresholds and limits.
Read the postAjay Patel
Cloudflare Clef vs Clef-flash vs Clef-omni: which model gets each decision
Cloudflare Clef vs Clef-flash vs Clef-omni, picked by input type, hosted context window and cost per million decisions, and where Clef-omni scores lower.
Read the postIshan Chavda
What are Web Vitals, how they work and why they matter for web performance
What are Web Vitals? LCP, INP and CLS defined with Google's thresholds, the 75th percentile rule, lab versus field data and one site's real numbers.
Read the postAjay Patel
Build or buy PIM for ecommerce: what a PIM is, when you need one and how to choose
Build or buy PIM for ecommerce? Test your catalogue against platform limits, then compare SaaS, open core, a custom build and no PIM on the same axes.
Read the postAjay Patel
How to hire UI/UX product designers and test the work before you commit
How to hire UI/UX product designers when you cannot judge design: write pass lines first, read portfolios for decisions, then run a short paid pilot.
Read the post










