Recent work you can inspect
I contribute ongoing performance improvements to established open-source Rails applications. Selected contributions reviewed and merged upstream:
- Gumroad Products API: co-authored batch-loading work that reduced SQL events from 103 to 23 and mean request duration by 40.4% in a ten-product benchmark.
- Discourse admin active-users endpoint: replaced per-user 2FA checks with batched queries, reducing mean request time by 66.5% in a local production-mode benchmark.
- OpenProject Work Packages API: eliminated 762 repeated queries and reduced median server duration by 55.7% in the documented benchmark.
Each case study includes the problem, my contribution, validation, and the upstream pull request. These are open-source contributions, not client engagements; the benchmark figures are not measurements of those projects’ live production traffic.
For a client production result, see how I reduced a Rails dashboard’s load time from over two seconds to under 200ms.
My background
I started as a full-stack developer and moved through software engineering, consulting, and technical leadership. Most of my experience comes from production SaaS applications where reliability, performance, security, and maintainability matter.
My work has included:
- Employee-compliance and vendor-due-diligence systems
- Software supply-chain compliance products
- Learning tools, AI-assisted features, and speech-enabled functionality
- Complex APIs, integrations, background processes, and data workflows
- Performance improvements for slow pages, database queries, crawlers, and memory-intensive jobs
- Product and infrastructure work on a SaaS platform serving more than one million users
- Engineering leadership, architecture, and code review
I have worked across education technology, financial compliance, cybersecurity, marketing technology, infrastructure automation, and SaaS. That variety taught me how to enter an unfamiliar domain, understand what matters, and improve a system without assuming it should be rewritten.
Problems I can help solve
AI-assisted workflows and product features
I help teams turn a repeatable research or operational process into a focused, maintainable tool. I can also add AI-assisted features to an existing product, connecting them to the application’s current data, permissions, background jobs, APIs, and business rules.
The aim is not to add an agent where a script or ordinary application would be better. I identify the part that benefits from model reasoning, keep deterministic rules in code, and leave consequential actions under appropriate human control.
Rails and production SaaS
Ruby on Rails and PostgreSQL remain my deepest application stack. I can join an established product to diagnose a problem, deliver a defined improvement, or provide expert engineering support.
Common engagements include:
- Rails, PostgreSQL, and Sidekiq performance improvements
- Rails upgrades and legacy application modernization
- APIs, third-party integrations, and background processing
- AI-assisted features inside an existing Rails product
- Difficult production issues and fragile operational processes
JavaScript and TypeScript support the product layer; Python supports workflow tooling, document processing, model integrations, and data analysis. I also have experience with GraphQL, AWS, CI/CD, and application performance engineering.
How I work
You work directly with me, from the initial investigation through implementation and handoff. I establish a baseline, explain the trade-offs, and deliver reviewable changes with tests and documentation. I use AI-assisted development to accelerate the work while remaining responsible for understanding and validating the result.
I am based in Kerala, India, and work remotely with international teams. Start with a focused Rails performance engagement or bring me in for a defined implementation period. We agree on scope, access, and working arrangements before starting.