103 SQL events reduced to 23.
Co-authored a Products API batch-loading fix. Mean request duration fell 40.4% in a ten-product benchmark, while preserving the response contract.
Read the measurements View merged pull requestHaseeb Annadamban · Senior Rails engineer
I help SaaS teams resolve performance bottlenecks, improve established Rails applications, and ship practical AI features. Ten years of production experience, backed by client results and merged contributions to Gumroad, Discourse, and OpenProject.
Available for remote consulting and contract work
01 Open-source performance improvements Merged upstream
Performance fixes for Gumroad, Discourse, and OpenProject: measured bottlenecks, regression tests, and changes reviewed and merged by maintainers. This is open-source contribution work, not a claim that these projects are clients.
Co-authored a Products API batch-loading fix. Mean request duration fell 40.4% in a ten-product benchmark, while preserving the response contract.
Read the measurements View merged pull requestReplaced per-user two-factor authentication checks with two batched queries. Measured locally in production mode with 75 active users and empty 2FA data.
Read the measurements View merged pull requestRemoved repeated type-variant lookups from the Work Packages API. Median server duration fell 55.7% in the documented benchmark.
Read the measurements View merged pull requestBenchmark results describe the workloads documented in each case study, not measured changes to these projects’ live production traffic. Your application’s results will depend on its own bottlenecks.
Explore all case studies02 Latest case studies Problems, approach, results
The latest detailed accounts of performance problems, the engineering approach used to address them, and the measured outcome.
An open source Rails performance optimization that made role checks association-aware across Forem’s admin member list, CSV export, async user payload, and reCAPTCHA service.
View case studyAn open source Rails performance optimization that replaced per-article analytics recomputation with bulk SQL aggregation, cutting median backfill time by 90.1%.
View case studyA co-authored open source Rails performance fix that batch-loaded product serialization data, cutting mean response time by 40.4% and SQL events by 77.7%.
View case study03 Ways to work together A clear first step
Bring me a slow application, a difficult backend project, or a workflow worth automating. We agree on the scope, the evidence of success, and the handoff before implementation begins.
For SaaS teams facing slow requests, expensive queries, unreliable jobs, or difficult legacy code. Start with a performance engagement, or bring me into your team for a defined delivery milestone.
For teams with a repeatable research or operational process that takes too much manual effort. Start with one bounded workflow, connected to the tools and data your team already uses.
Not sure where to start? The initial fit discussion is free. We’ll clarify the problem, available access, and scope before any paid work begins.
04 Production work
Beyond open source, I have delivered production work across education, compliance, and SaaS—from faster dashboards to integrations and infrastructure upgrades.
Built spaced-repetition, pronunciation, and reading tools for a nonprofit serving underserved students—including work families credited with improving their children’s grades.
Product engineering · Applied AIBuilt employee-compliance and vendor-due-diligence systems, plus meeting transcription across Google Meet, Zoom, and Microsoft Teams.
Workflow systems · IntegrationsCut dashboard load time by over 90%, removed crawler memory bloat, and improved sitemap processing speed and worker reliability. Read the production case study.
Rails · PostgreSQL · PerformanceImproved page speed, rebuilt a CMS rendering system, raised SEO performance, upgraded critical infrastructure, and delivered new integrations.
Architecture · Modernization05 Testimonials Client perspective
Feedback from founders who have trusted me with their products, workflows, and production systems.
06 AI work in public
I use AI to accelerate the work, not to skip understanding or validation. My open-source tools show how I build repeatable workflows with verifiable outputs, approval gates, and documented handoffs. These are public engineering projects, not claimed client deployments.
A multi-stage customer-research method packaged as a reusable skill with sourced evidence, structured artifacts, validation scripts, and an approval checkpoint.
Turns repeatable agent work into versioned runbooks with approval gates, checkpoints, output verification, and evidence-based revisions.
Deploys Pi to SSH-accessible Linux servers for terminal, Telegram, or Slack access without public ports, TLS, webhooks, or manual setup.

07 Engineering perspective
I work directly with founders and engineering teams to understand unfamiliar systems, make the trade-offs explicit, and deliver reviewable changes. Rails and PostgreSQL are my deepest stack; Python and TypeScript support my AI and workflow tooling. Based in Kerala, India, I work remotely with international clients.
Founders and technical leaders at Luthor, Interlynk, and Learn To Be describe my ownership, product judgment, and delivery in their client testimonials.
More about my backgroundFocused engagements · Senior contract engineering
Tell me what is happening, why it matters, and your timeline. We’ll identify a sensible first engagement before committing to implementation.
Discuss your project