Suraj's Profound AI Rep
Suraj builds backend systems that hold up in production and stay flexible as the product evolves.
Suraj Nanda
Edges
Owning Backend Services End to End
Suraj takes backend services from design through production and holds them there. He does not hand off at the boundary between build and live; he owns both. At mPokket, he designed the Astro Lab service from scratch, led a team of four through the full SDLC, made the database architecture call, and delivered within the planned timeline. At Acko, he owns production reliability for live insurance systems, running root-cause analysis and shipping fixes under pressure. His pattern is to understand the business requirement first, make the architectural decisions that give the system room to evolve, and then stay accountable for what happens after it ships. For a team that needs someone who will not drop the ball between build and production, this is the throughline across his career.
Spotted in 3 Stories
Building Event-Driven Reliability Systems
Suraj builds backend systems where events cannot be lost and consistency cannot be assumed. His instinct is to design for failure modes before they surface in production. At mPokket, he built a Kafka consumer handling tens of thousands of daily calls, with an idempotency layer he designed to prevent duplicate event processing from corrupting user data. At Acko, he identified a dual-write gap and implemented the outbox pattern to guarantee at-least-once delivery across services. In both cases, the work was not just about making the happy path work. It was about making the system correct when things go wrong: Kafka goes down, events arrive twice, downstream services are unavailable. For teams building on Kafka or any event-driven architecture, his record shows he understands the failure modes and designs for them.
Spotted in 2 Stories
Architectural Decision-Making Under Constraints
Suraj makes backend architectural decisions grounded in the actual constraints of the product, not just the ideal design. He evaluates tradeoffs against business timelines, expected scale, and the team's ability to maintain what gets built. At mPokket, he chose NoSQL for the Astro Lab service because the schema was variable and product requirements were expected to shift. He chose the company's existing Java and Spring Boot stack to avoid unnecessary exploration. Both calls were deliberate and reasoned. At Acko, he identified the outbox pattern as the right solution to a dual-write problem, understanding both the failure mode and the mechanism needed to address it. His approach to service boundaries follows the same logic: start with business domains and data ownership, then let the architecture follow.
Spotted in 2 Stories