Nishtha Gupta

Nishtha Gupta

software engineer · college park, maryland

I design event-driven systems on AWS

Every service choice answers a specific access pattern.

Seeking Summer 2027 software internships

View projects Get in touch GitHub LinkedIn

  • 85ms → 12ms

    p99 verification latency, a 7× improvement, benchmarked in CloudWatch

  • 2 hrs → 15 min

    regression suite runtime after automating 100+ tests at TIAA

  • 100+ hrs

    engineering time recovered per sprint for a 5-engineer delivery group

climatewatch — ingest to query

Data flows from Kinesis into DynamoDB for key-value lookups and into S3 for columnar scans via Athena, with EventBridge orchestrating. Kinesis stream buffering EventBridge orchestration DynamoDB key-value lookup S3 columnar store Athena scan on read
Each managed service answers one access pattern — stream buffering, key-value lookup, object storage, columnar analytics — which keeps ingestion, storage, and analytical workloads independently operable.

about

Software engineer with 2+ years building backend systems and cloud infrastructure, now doing an M.Eng. in Cloud Engineering at the University of Maryland. Python, Java, REST APIs, AWS, Terraform, Kubernetes, CI/CD, observability, and both relational and NoSQL stores.

I led backend development for customer-facing products, and a good share of my production experience is the unglamorous half of distributed systems: tracing participant data across MuleSoft APIs, MDM, Snowflake, and Salesforce to isolate where records diverged. That work is why I instrument observability while building rather than after something breaks.

education

  • M.Eng. Cloud Engineering — University of Maryland, College Park Jan 2026 – Dec 2027

    Advanced Networking · Databases · Cloud Computing · Linux System Administration · Virtualization and Container Technology

  • B.E. Computer Science and Engineering — University of Mumbai Aug 2019 – May 2023

    Operating Systems · Computer Networks · Distributed Systems · Database Systems · DSA

projects

Systems, and the decisions behind them

Lottery Verification Platform

Infrastructure and Networking Lead · 2026

A verification service where I owned the network and infrastructure layer, then kept extending it after the team build ended.

  • Configured AWS networking and container infrastructure across VPC, load balancing, ECS, and security groups, defining the deployment path and service-level access controls.
  • Devised an SQS dead-letter queue workflow to isolate verification requests failing during burst traffic, preserving failed work for inspection and controlled retry instead of losing it silently.
  • Reduced p99 latency from 85ms to 12ms — a 7× improvement — by introducing ElastiCache Redis for critical verification data, cutting database utilisation by 25%.
  • Benchmarked latency and database utilisation in CloudWatch before and after the cache, so the improvement is a measured number rather than an assumption.
AWSECSVPCSQSRedis

source ↗

ClimateWatch

2026

A streaming climate-data pipeline that ingests, persists, and queries records through event-driven workflows.

  • Orchestrated the pipeline across Kinesis, DynamoDB, S3, Athena, and EventBridge.
  • Designed service access patterns around stream buffering, key-value lookup, object storage, and columnar analytics — mapping each workload to Kinesis, DynamoDB, S3, and Athena respectively.
  • Separated ingestion, storage, and analytical workloads through event-driven components, so each can be operated and scaled independently.
PythonKinesisDynamoDBS3AthenaEventBridge

source ↗

AgriSense Agent

2025

A retrieval-augmented agricultural advisory agent that generates attributed crop recommendations grounded in retrieved evidence rather than unsourced text.

  • Architected the agent in Python to retrieve region-specific agricultural sources and attribute every recommendation to them.
  • Deployed the RAG pipeline on AWS with application and retrieval observability, enabling end-to-end diagnosis of latency across retrieval and inference stages separately.
  • Formulated a fixed evaluation set for repeatable retrieval-quality testing, making agent behaviour comparable across iterations instead of judged by inspection.
PythonRAGAWSObservability

source ↗

also built

Bedrock

academic · 2026

Modular Terraform infrastructure deploying a 3-tier AWS architecture entirely as code.

  • Five modules — networking, security, compute, storage, monitoring — deploying ALB → ASG web tier → RDS, with databases isolated in private subnets behind a NAT Gateway.
  • Conditional resource counts let one shared codebase support dev, staging, and prod without duplication.
TerraformEC2ALBASGRDS

Surge

academic · 2026

A message-driven pipeline that autoscales EKS worker pods against queue depth.

  • SNS → SQS → EKS workers, with a KEDA ScaledObject scaling pods from 1 to 10 on queue depth, polling every 15 seconds.
  • EKS provisioned via eksctl, KEDA deployed through Helm, IAM inline policies scoped to least privilege; scale-out and scale-in validated under a 100-message load test.
EKSKEDASNSSQSHelm

Polyglot Healthcare System

2026

A healthcare records system using three databases, each chosen for the queries it is actually good at.

  • Python model and repository layer over PostgreSQL, isolating SQL from business logic so clinical workflows compose without touching driver details.
  • Neo4j catalog spanning diseases, medications, symptoms, and lab tests, enabling traversal-based drug-interaction checks that gate prescriptions before they reach the system of record.
PostgreSQLMongoDBNeo4jDocker

source ↗

Driver Drowsiness Detection

published · 2023

Real-time driver-fatigue detection on Raspberry Pi, without specialised in-vehicle hardware.

  • Detection and alert workflow built around real-time facial and eye-state analysis using OpenCV and Dlib.
  • Findings published and presented at Hinweis ASIT 2023.
OpenCVDlibRaspberry Pi

skills

Core

What I will defend in depth, unprompted.

languages

PythonJavaC++JavaScriptSQLBash

cloud and infrastructure

EC2VPCECSEKSS3RDSSQSSNSKinesisEventBridgeAthenaElastiCacheCloudWatchIAMTerraformDockerKubernetesLinux

backend and apis

Node.jsExpressSpring BootDjangoDjango REST FrameworkFlaskREST API designEvent-driven architecture

databases and data

MySQLMariaDBPostgreSQLDynamoDBSnowflakeRedisMongoDBNeo4j

software engineering

CI/CDAutomated testingSelenium WebDriverTestNGCopadoObservabilityDistributed systemsAuth & authorizationDSA

tools

GitJiraSplunkHelmeksctl

experience

  1. Graduate Assistant

    Jan 2026 — Present

    University of Maryland, College Park · College Park, MD

  2. Software Analyst

    Jul 2023 — Nov 2025

    TIAA · Mumbai, India

    Best Fresher of the Year, 2024

    • Automated 100+ regression and integration tests with Selenium WebDriver, Copado, and TestNG, cutting suite runtime from 2 hours to under 15 minutes.
    • Traced participant and institutional data across MuleSoft APIs, MDM, Snowflake, and Salesforce, isolating transformation and synchronisation issues across customer-facing workflows.
    • Recovered 100+ engineering hours per sprint for a 5-engineer delivery group by eliminating repetitive manual validation.
    • Investigated MDM defects causing duplicate and mismatched profiles, reducing record-related service escalations by 60%.
    • Validated MuleSoft API data loads and transformations, catching 5–7 critical import and data-quality issues per release cycle before production.
    • Validated the Siebel-to-Salesforce migration across millions of records and 75+ legacy entities, surfacing 150+ field-level discrepancies before cutover.
    • Reconciled Salesforce operational data against Snowflake analytical datasets, validating alignment with the system of record.
    • Designed a release-monitoring dashboard tracking test phases, defect status, and validation progress across recurring deployment cycles.

  3. Backend Team Lead

    Nov 2021 — Jan 2023

    SHRAM · Mumbai, India

    • Led a 3-engineer backend team within a 9-person product organisation, setting technical direction and implementation standards for 15–20 customer teams.
    • Owned the module-by-module migration from Java/Spring Boot to Node.js, preserving REST API contracts while the legacy system stayed operational, and executed a controlled ~2-hour production cutover.
    • Optimised task-assignment and delegation workflows by indexing lookup paths and consolidating repeated queries, reducing database round trips from up to 3 per request to 1.
    • Built a performance analytics service and REST API layer consumed by web dashboards, improving dashboard load times by 30%.
    • Designed Firebase notifications for mobile and web clients, supporting ~70 daily and ~250 peak notifications.
    • Established API design guidelines and code-review practices, reducing frontend/backend integration rework, and mentored 3 junior developers.

    Part-time, promoted from Backend Developer. Two stints — Nov 2021 to May 2022 and Aug 2022 to Jan 2023 — held alongside full-time undergraduate coursework. Gamified B2B productivity platform.

  4. Student Intern

    Jun 2022 — Jul 2022

    TIAA · Pune, India

    • Developed RESTful APIs with Django REST Framework for an AI/NLP application, building Python backend services and integrating components end to end.
    • Built user-facing features with Django, HTML, CSS, and Bootstrap alongside cross-functional Agile teams.
    • Validated application behaviour across 3 device emulators spanning Android and iOS, and turned testing findings into usability recommendations that shipped.

publication

Driver Drowsiness Detection and Alert System

Published and presented at Hinweis ASIT 2023 — Second International Conference on Advances in Software Engineering and Information Technology.

leadership

  • NSS Lead — National Service Scheme

    Led nonprofit education and outreach initiatives reaching 500+ underprivileged children.

  • Women in Tech Lead — Google Developer Student Clubs, RAIT

    Organised technical events and mentored students.

  • Intern Lead — SDG Cell, RAIT

    Mentored 27 interns and coordinated stakeholders across project delivery.

contact

Get in touch