Use AI to propose and implement refactors ranging from local function cleanup to repo-aware coordinated code changes across larger codebases, including structural and architectural changes, API contract alignment, and other multi-file edits that benefit from repository context.
The engineer decides the refactor strategy, reviews diffs, runs builds and tests, validates contract and pipeline impacts, confirms behavior and performance remain correct, and ensures smaller cleanups do not introduce regressions; AI suggestions must be verified before they ship.
AI rewrites a long function into smaller helper methods with clearer names and responsibilities while keeping the original behavior intact.
Cursor / ClaudeAI updates call sites across the repo to match a renamed API or signature change while keeping formatting and imports consistent.
Claude Code / Sourcegraph CodyAI identifies duplicated logic across services and proposes a shared module/library extraction with incremental PR-sized changes.
Claude Code / GitHub Copilot ChatI help companies navigate AI adoption -- from strategy to production. Whether you are building your first LLM-powered feature or scaling an agentic system, I can help you get it right.
Design and build LLM-powered products and agentic systems
Go from idea to production with a clear implementation roadmap
Build AI with human-in-the-loop in regulated environments
I used AI every day for 30 days as a backend engineer.
Not for side projects or experiments — but in real work: debugging issues, writing code, understanding systems, and reviewing logic.
Some things became 10× faster.
Others actually made me worse.
Here’s what actually changed.
There is no shortage of AI tools aimed at developers right now: chat assistants, IDE completions, agents that promise to run your tests, and new products every month with overlapping features. I am not going to argue which one is “best.” Instead, here is what I am actually using today on hobby code: ChatGPT for quick, low-context questions, Cursor when the work needs my repository in the loop, and GitHub Copilot for fast inline help while I type. That trio might change, but it reflects how I have learned to spend money and attention in 2026.
The through-line is simple: match the tool to how much context the problem needs. That stops me from dumping half a repo into a browser tab for a vague design question, or firing up an editor assistant when I only wanted a two-paragraph explanation of something I could read in the docs.
This is not a product review. It is a snapshot of how I work, using the projects on my projects page as concrete examples.
I've been using Claude Code daily for 6+ months. It's replaced most of my routine work:
• To understand code base - yes true this is game changing for me. its creates nice tech documents.
• Boilerplate generation
• Refactoring large codebases
• Debugging edge cases
• Writing tests & automation scripts - top notch in UT's
It outperforms Cursor and GitHub Copilot for deep reasoning and architectural changes. Period.
⚠️ BUT — always validate. Review the code. Run your tests. You're still the final gatekeeper. AI hallucinates less here, but "less" ≠ "never."