Shreyaan Seth
01

Case studies

The decisions behind the systems, written out in full.

Four production problems, explained the way I would in a design review: what broke, the options I rejected, what changed, and the result.

  1. Agent infrastructure

    When an agent boundary stopped paying for itself

    A production agent runtime sent its execution state to a second LLM just to write the final answer. I wrote the RFC to fold synthesis into the primary runtime and replace agent-level debugging with step-level tracing.

    • Agent runtimes
    • LLM orchestration
    • Latency optimization
    • Observability
    • Migration design
    • RFC ownership
    ~35s → ~13stotal response time in some cases

  2. Distributed systems

    Making long-running workflows survive process failure

    A deploy could delete a workflow in the middle of a two-day wait. I traced the problem to in-memory continuations and moved the runtime to persisted waits and duplicate-safe resume claims.

    • Durable execution
    • Workflow engines
    • Idempotency
    • Distributed systems
    • Failure semantics
    • PostgreSQL
    Deploy-safedurable waits, approvals, and duplicate-safe resume claims

  3. Multi-tenant security

    Moving tenant isolation from agent prompts into PostgreSQL RLS

    An AI SQL agent queried shared multi-tenant tables. I designed and implemented tenant isolation with PostgreSQL Row-Level Security, from authenticated context through connection-pool safety.

    • PostgreSQL RLS
    • Multi-tenancy
    • AI security
    • Authorization
    • Row-level security
    • Agent design
    Prompt → RLSauthorization enforced by PostgreSQL, not by generated SQL

  4. Cloud infrastructure

    Right-sizing an overprovisioned production PostgreSQL workload

    A production database cost about $988 a month while averaging 0.8% CPU and barely using its 30,000 provisioned IOPS. I right-sized compute and storage performance, bringing the RDS bill down to about $320.

    • AWS RDS
    • PostgreSQL
    • Capacity planning
    • Cloud cost optimization
    • Production risk
    • Graviton
    $988 → $320/moobserved RDS spend, ~68% lower

Currently open to the right problem

Want to talk through the details?

I'm happy to walk through the architecture, the tradeoffs, and the parts that did not work.

shreyaans20@gmail.com