Senior Software Engineer · Gurgaon
Ten years of platforms with real users on the other end.
Amazon, 2016 to 2026 — Bengaluru then Seattle, finishing as a Senior SDE. Three domains, one habit: own the thing end to end, including the pager.
The numbers that hold up
- 250ms → 40ms API latency
- Seller rewards, after an Oracle-to-DynamoDB migration.
- 20 → 5 tables
- Relational schema redesigned around O(1) access patterns.
- 66% TPS reduction
- Downstream load, from a typed eligibility API with caching.
- 32 → 2 min UI test time
- A 94% cut across eight languages.
- 12 countries launched
- One EMEA expansion, four phased launches, on schedule.
- 10 days → 12 hrs incident detection
- Per-template bounce metrics replacing a silent failure mode.
Four systems
The work I would most want to be asked about. Each of these is drawn rather than described, because the shape is the argument.
2018–2019
One service where there had been ten scripts
Seller notifications were scattered across a decade of ad-hoc jobs. Ten-plus script-based clients each tracked their own success, errors and retries, so the organisation had no single answer to the only question that mattered: did the seller actually get it? I proposed the consolidation, got approval, and built one platform across WhatsApp, SMS and email.
Unified seller notification platform. Two ingress paths, one service, three channels.
What was hard The return path. Provider status callbacks reconcile through a Lambda into DynamoDB under concatenated keys, so delivery state is queryable in O(1) instead of living in ten scripts’ private logs. Delivery tracking became a first-class concern rather than something bolted onto each sender.
2022–2026
A career-profile platform, from inception
A gateway and its data layer, built from nothing on AppSync, Lambda, DynamoDB and CDK. I wrote the low-level design — service decomposition, data flow, caching — and the GraphQL schema-driven patterns the team went on to adopt. It ended up serving career-mobility features across fourteen countries and forty-odd locales.
Career-profile platform. Schema-driven gateway, single-table store, event fan-out.
What was hard Choosing GraphQL over REST for clients whose query shapes kept changing, then keeping a single-table DynamoDB design honest as access patterns multiplied. Change streams fan out through SNS so downstream consumers never poll.
2025
Configuration that stops drifting
Two independently-owned services held overlapping configuration and drifted apart by hand. I designed an event-driven sync: a config object lands in S3 and flows through SNS, a FIFO queue, schema validation and a manual approval gate before it is written.
Cross-service configuration sync. Ordered, validated, gated, idempotent.
What was hard Ordering and replay. FIFO delivery keeps changes in sequence, and the final write is idempotent, so a redelivery cannot corrupt the target. The approval gate keeps a human in the loop without putting one in the critical path.
2024–2025
Finding failures that were failing silently
Email delivery failures were going unnoticed for over a week. I integrated a preference service to stop the bleeding, then built monitoring so the next one would surface in hours.
Bounce monitoring. Per-template metrics turn an invisible failure into a paging one.
What was hard Granularity. Aggregate bounce rates hid the problem — a single misbehaving template disappears into a healthy overall number. Per-template metrics, composite alarms and a daily report cut detection from ten days to under twelve hours.
The arc
- 2024–2026Seattle · career mobility
Senior Software Development Engineer
Led a twelve-country EMEA expansion across four phased launches, coordinating dependency teams in several organisations, then wrote the runbook that cut each later country launch by roughly 60%. Took a core service to full continuous-delivery certification in a week against a three-week estimate.
- 2021–2024Seattle · across several teams
Software Development Engineer II
Built the career-profile gateway and data layer from inception. Built a configurable matching engine on a graph model, so product managers could tune how candidates rank against roles without a code change — plus time-boxed delegation with TTL-based access revocation. Earlier in the stretch, a capacity-reservation system for a logistics platform and claims auto-resolution across multi-country launches.
- 2018–2021Bengaluru · seller experience
Software Development Engineer II
Proposed and built the unified notification platform. Migrated a rewards backend from Oracle to DynamoDB. Delivered India’s GST tax-reform integration across six seller-facing services, ahead of the government deadline. First junior engineer on the team made a code-review bar raiser.
- 2016–2018Bengaluru · registration & onboarding
Software Development Engineer
Built India’s seller post-launch onboarding portal from scratch on a template pattern, and the event pipeline behind it — first on the team to introduce gradual-rollout experimentation.
How I work
- Alarms that page a human
- An audit across five services found four of five composite alarms routing to nobody. An alarm nobody receives is worse than no alarm, because it buys false confidence.
- Incident reviews written honestly
- Two production correction-of-error documents, both with the awkward parts left in. A review that protects someone’s feelings teaches nothing.
- Tech debt quantified, not complained about
- I catalogued open debt across nearly two years of tickets and used blocking-chain reasoning to show one migration unblocking three initiatives at once. That turns an argument into a decision.
- Pipelines that refuse bad changes
- Full continuous-delivery certification on a core service in a week — reviewer gates, bake times, auto-rollback monitors — then made reusable so the rest of the team inherited it.
Multiplying others
The part that compounds. Most of what I am proud of from the last few years is not code I wrote.
- Around 300 code reviews a year, with feedback on race conditions, internationalisation regressions, monitoring gaps and security risk — not just style.
- Ran a promotion assessment across nine stakeholders including three principal engineers, and wrote promotion feedback for several teammates.
- Interviewed for SDE and campus hiring across multiple quarters — coding, system design and behavioural.
- Built the team wiki from scratch, replacing tribal knowledge with something a new joiner could actually read.
- Drove AI-assisted development across engineering, product and data teams, including the guardrails that stop it going wrong.
What I am looking for
Senior or Staff engineering roles, based in India, remote or hybrid. Distributed systems, platform work, or anything where owning the whole lifecycle is the point. Applying from December 2026.