Available for impactful work Sunnyvale, CA · PT

Shweta Mishra.

Full-stack software engineer building large-scale distributed systems, high-throughput multithreaded services, and production AI agents for Fortune 1 platforms.

4+years shipping
production systems
200K+sellers on platforms
I help build
420M+listings powered
by my services
3Walmart Global Tech
awards in 2024
scroll
01

About

I architect systems that move real money, real shipments, and real decisions — at the scale of one of the world's largest retailers.

At Walmart Global Tech, I work on Walmart Marketplace — the platform hosting 200,000+ third-party sellers and 420M+ listings (~95% of Walmart.com's catalog) — and the Enterprise Data Platform powering metadata for the entire company.

Lately I've been pushing into agentic AI: I pioneered and shipped an autonomous SRE agent (ReAct on LangGraph) that ingests alerts, runs investigative tool-calls, cross-references runbooks via RAG, and produces structured RCA reports without on-call intervention.

I hold an MS in Computer Science from Boston University. Before grad school, I shipped CRM and API platforms at Accenture.

02

Career Journey

  1. Jan 2024 — Present Sunnyvale, CA

    Software Engineer 3 @ Walmart Global Tech

    Walmart Marketplace

    200K+ sellers · 420M+ listings · ~95% of Walmart.com catalog

    • Architected & launched End-of-Day Scan Form (EODSF) from zero to production — consolidated shipment manifest letting sellers hand off all daily shipments to carriers in a single scan. Owned GraphQL schema, REST services, and end-to-end functional/load/contract testing. 30% lift in seller satisfaction; full SLA compliance through peak holiday volume.
    • Designed & delivered the Promotions Platform — real-time promotional spend & attribution across Marketplace. Engineered the attribution engine in Java with ForkJoinPool, CompletableFuture pipelines, and lock-free concurrent data structures to hold sub-second latency at production scale.
    • Authored concurrent processing modules for high-volume event fan-out with rigorous thread-safety, backpressure, and memory-bounded queues — patterns adopted by adjacent teams as an internal reference implementation.
    • Pioneered an autonomous AI SRE agent for Marketplace operational excellence — a background ReAct-pattern agent on LangGraph (Python) that ingests alerts and logs from OpenObserve, runs investigation tool-calls (log search, metric correlation, deployment diff, dependency trace), and produces structured RCA reports without on-call intervention.
    • Engineered a RAG pipeline over historical incidents and runbooks using vector embeddings — agent recognizes recurring failure signatures and cites precedent. Compressed MTTU on recurring incidents from hours to minutes.
    • Designed the agent's tool-building framework (deterministic tool schemas, retry/backoff, token-budget guardrails, grounding checks) so other engineers can onboard investigation capabilities without touching the core graph.
    Enterprise Data Platform (EDP)

    Cloud & Data Platforms — enterprise metadata infrastructure

    • Owned backend development of Walmart Data Platform (WDP) using Spring Boot, Java, and Google Bigtable — centralized metadata infrastructure supporting enterprise-wide decision-making. 20% improvement in downstream consumer satisfaction.
    • Led integration of Erwin Data Modeler metadata into WDP over a Kafka streaming pipeline — exactly-once delivery, schema evolution, idempotent consumers across distributed services. Recognized by leadership as a critical EDP feature.
    • Delivered Erwin Model Enhancements & UI improvements to the Data Modeler tool used by internal data modelers across Walmart Global Tech.
    • Drove a data-quality initiative in Elasticsearch that eliminated 34% of redundant indexed documents, measurably improving search relevance and query latency.
    JavaSpring BootPythonLangGraph LangChainKafkaGoogle Bigtable ElasticsearchGraphQLReact OpenObserveVector DBs
  2. Jul 2018 — May 2021 Mumbai, India

    Software Engineer · Full Stack @ Accenture

    • Delivered a Spring Boot CRM platform letting administrators manage customer profiles and purchase history, unlocking data-driven decision-making for enterprise e-commerce clients.
    • Designed & implemented RESTful APIs using Spring Web MVC and Spring Boot — flexible, scalable access to customer data.
    • Implemented unit & integration tests with JUnit and Mockito and wired them into the CI/CD pipeline. 15% reduction in errors, improved application reliability.
    JavaSpring BootSpring MVC RESTJUnitMockitoCI/CD
  3. Jan 2022 — May 2023 Boston, MA

    Master of Science · Computer Science @ Boston University

    Built Tixfix — a ticketing service with a Node.js + TypeScript backend and React frontend featuring secure auth, real-time updates, and customizable ticket workflows.

  4. Jul 2014 — May 2018 India

    Bachelor of Technology · Computer Science @ Mody University

03

Personal Projects

github.com/smshweta ↗

Live from my GitHub — fetched in real time, sorted by recent activity.

04

Skills & Stack

LanguagesJava · Python · TypeScript · JavaScript · HTML/CSS
FrameworksSpring Boot · Spring MVC · Node.js · React · LangGraph · LangChain
AI / MLReAct Agents · Agentic Workflows · RAG · Embeddings · Vector Search · Tool-Calling
ConcurrencyJava Multithreading · ForkJoinPool · CompletableFuture · Concurrent Collections
DistributedKafka · Kubernetes · Docker · Event-Driven Architecture
DatabasesMongoDB · PostgreSQL · MySQL · Google Bigtable · Elasticsearch · Vector DBs
Data ModelingErwin Data Modeler · Metadata Management · Schema Evolution
ObservabilityOpenObserve · Structured Logging · Automated RCA Pipelines
Cloud / ToolsGCP · Azure · Git · CI/CD
MethodologiesSDLC · Agile · TDD
05

Let's talk

I'm always up for conversations about distributed systems, agentic AI, and platforms operating at scale.