Software Engineer Co-op at Crewasis, Jan–May 2026 — open to backend/cloud roles from Jan 2027

Backend & Cloud
Engineer

MS in Software Engineering @ Northeastern (3.9 GPA), graduating Dec 2026. Building production APIs, cloud infrastructure, and CI/CD pipelines. Most recently a Software Engineer Co-op at Crewasis in New York.

View Projects
GitHubGet in Touch
3.9
MS GPA — Northeastern
Dec '26
Graduating
Jan '27
Available Full-Time
Scroll

About

Backend engineer who ships.

I'm a Master's student at Northeastern University finishing my MS in Software Engineering (3.9 GPA), graduating December 2026. Most recently I spent five months in New York as a Software Engineer Co-op at Crewasis — writing Django APIs, optimizing PostgreSQL queries, provisioning AWS infrastructure, and shipping weekly. Before that, two internships at Tatvasoft in Ahmedabad building REST APIs in C#/ASP.NET Core, and nine months as a research assistant at Nirma University working on reinforcement learning and audio classification. I focus on backend services, cloud infrastructure, and the CI/CD that ties them together.

Backend Services

Building REST APIs with Django, Spring Boot, FastAPI, and ASP.NET Core. Query optimization, connection pooling, and caching with Redis for low-latency responses.

Cloud & Infrastructure

AWS (EC2, RDS, S3, Lambda, SNS), Terraform IaC, Docker containers, and GitHub Actions CI/CD. Infrastructure that's reproducible and version-controlled.

Databases & Storage

PostgreSQL, MySQL, MongoDB, Redis. Schema design, indexing strategies, query plan analysis, and caching layers for production workloads.

Experience

Where I've worked.

University research in India, then two internships at the same company there, then a co-op in New York. Listed newest first.

ResearchIndia · 9 moIndustryIndia · 3 + 5 moCo-opNew York · 5 mo

Crewasis

New York, NYUnited States

Software Engineer Co-op

Jan 2026 — May 20265 months
Shipped backend features in Django on a weekly release cadence, holding test coverage above 90% on the modules I wrote
Containerized the services with Docker and moved builds and deploys into GitHub Actions, replacing a manual release process
Cut p95 latency ~35% (measured in Django Debug Toolbar) and query volume ~30% (measured with pg_stat_statements) through PostgreSQL query-plan optimization and a Redis cache layer
Provisioned the AWS footprint the application runs on — EC2, RDS, S3 and Lambda
PythonDjangoAWSDockerPostgreSQLRedisGitHub Actions

Tatvasoft

2 roles
Ahmedabad, IndiaIndia

Two separate internships — invited back for a second, longer stint the following year.

Software Developer Intern

Jan 2024 — May 20245 months
Built RESTful API endpoints in C#/ASP.NET Core with Entity Framework against an existing PostgreSQL schema, load-tested to a 100+ concurrent-user target at p95 under 500ms
Cut query time ~30% on the slowest endpoints through PostgreSQL indexing and query-plan analysis
Set up CI/CD in Azure DevOps, with xUnit integration tests running on every pipeline execution
C#ASP.NET CorePostgreSQLAzure DevOpsEntity FrameworkxUnit

Software Developer Intern

Jun 2023 — Aug 20233 months
Built REST API endpoints for an MVC application
Implemented business-logic layers against the existing data-access code
C#ASP.NET MVCREST APIs

Nirma University

Ahmedabad, IndiaIndia

Research assistant under Prof. Anuja Nair, across two projects.

Research Assistant

Mar 2023 — Dec 20239 months
Autonomous vehicle collision avoidance: built the traffic simulation environment in SUMO and trained a reinforcement-learning policy for the avoidance behaviour
Lung sound disease detection: six-class classification over the ICBHI 2017 respiratory sound dataset — Butterworth bandpass filtering at 250–2000 Hz, augmentation by time stretching, pitch shifting and audio shifting, mel-spectrogram features, and EfficientNet-B0 with an added attention mechanism
Reached ~92% accuracy on a cycle-level split, which I'd now call optimistic: ICBHI has 126 patients, so splitting at cycle level puts the same patient on both sides, and augmented copies of one clip land in train and test together. A patient-level split is the correct protocol for this dataset and would score lower
Lung Sound DetectionCollision Avoidance
PythonPyTorchEfficientNetSUMOReinforcement LearningLibrosa

Education

Northeastern University

GPA: 3.9

Master of Science in Computer Software Engineering

Boston, MA · Sept 2024 — Dec 2026

Object-Oriented DesignData Structures & AlgorithmsCloud ComputingHigh-Performance Parallel ComputingData Science Engineering

Nirma University

Bachelor of Technology in Computer Science & Engineering

Ahmedabad, India · Oct 2020 — May 2024

Database Management SystemsOperating SystemsComputer NetworksDiscrete Mathematics

Projects

What I've built.

Backend services, cloud infrastructure, and distributed systems — each with real architecture decisions and measurable outcomes. Click any project for a full case study.

Featured Project

Distributed Task Queue

Celery + RabbitMQ — Circuit Breakers, Observability, Terraform

Production-grade distributed task queue processing asynchronous workloads (image transformation, PDF rendering, webhook dispatch) with FastAPI, Celery, and RabbitMQ. Includes fault-tolerant resilience patterns, full observability, and Terraform-based AWS deployment.

Every terminal failure recorded in an inspectable dead-letter state — no silent drops
Zero duplicate executions across 10K jobs under induced worker crashes
8 containerized services with Prometheus/Grafana observability
FastAPICeleryRabbitMQRedisPrometheusGrafanaDocker ComposeTerraformAWS ECS FargatePostgreSQL
CodeView Case Study
Featured Project

Real-Time AI Voice Agent

Sub-Second Phone Conversations — Streaming STT + LLM + TTS

Production voice agent that handles real phone calls with sub-second response latency. Composes Deepgram STT/TTS with an LLM router into a streaming pipeline so the agent starts replying before the caller has finished speaking.

Sub-second perceived latency via end-to-end streaming (no full STT→LLM→TTS wait)
Barge-in support — caller can interrupt the agent mid-sentence
Interrupt-aware router discards stale partial transcripts
Twilio Media Streams + WebSocket for full-duplex audio
PythonDeepgramTwilio Media StreamsWebSocketLLMStreaming
CodeView Case Study
Featured Project

Resilient Cloud Deployment Platform

Full AWS Stack — Terraform IaC, Multi-AZ, Blue-Green CI/CD

Production-grade cloud platform spanning 3 repositories: a FastAPI web service, Terraform IaC for the full AWS stack, and serverless Lambda functions for event-driven email verification. Designed to survive an AZ outage and ship via PR-triggered rolling deploys.

Multi-AZ VPC across 3 zones with Auto Scaling + ALB
Terraform IaC managing every resource across 3 repos
Zero-downtime CI/CD: PR validation → AMI bake → rolling deploy
Serverless email flow: SNS → Lambda → DynamoDB → SendGrid
AWSTerraformFastAPIPackerGitHub ActionsLambdaSNSDynamoDBRDS PostgreSQLS3
CodeView Case Study
Featured Project

Vehicle Rental Microservices Platform

Spring Boot — Concurrent Booking with Transactional Locks

Ride-booking platform built on Spring Boot microservices. The interesting problem: prevent double-booking when multiple users hit checkout for the same vehicle in the same second. Solved with row-level locking on availability + Redis caching for high-read paths.

Concurrent ride-booking via row-level locks — no double-bookings under contention
Sub-200ms reads via Redis caching of vehicle listings and sessions
JWT/RBAC isolating customer, driver, and admin permission sets
Bounded-context microservices: auth, bookings, payments, notifications
JavaSpring BootPostgreSQLRedisJWT/RBACNext.js
CodeView Case Study

Distributed Training Infrastructure

Multi-GPU Pipeline — PyTorch DDP, HDF5, ETL

High-performance distributed training system using PyTorch DDP across multiple GPUs with optimized ETL pipelines and HDF5 compression for large datasets.

PythonPyTorch DDPHDF5ETLWeights & Biases
Case study
Hackathon Winner

RareSense — Rare Disease Diagnostic Tool

Harvard Rare Disease Hackathon 2025

AI-powered diagnostic tool using phenotype-driven disease matching with the Human Phenotype Ontology (HPO) to help healthcare professionals identify rare genetic disorders.

PythonStreamlitHPOBioinformaticsData Science
Case study
Hackathon Winner

PaperVet — Research Paper Compliance AI

MINeD 2025 Hackathon · 3rd Place

Computer-vision pipeline that audits research papers against submission guidelines automatically — catching formatting and compliance issues before reviewers see them.

PythonOpenCVPyTesseractOCRComputer Vision+1
Case study

Air Chords

Vision-Based Musical Interface with Hand Gestures

Real-time computer vision application that transforms hand gestures into musical chords using OpenCV and MIDI synthesis.

PythonOpenCVMIDIComputer VisionReal-Time Processing
Case study
Research

Lung Sound Disease Detection

Attention over Pre-trained EfficientNet — ICBHI 2017

Six-class respiratory disease classification from lung auscultation audio, using mel-spectrogram features and an EfficientNet-B0 backbone with an added attention mechanism.

PythonPyTorchEfficientNet-B0AttentionLibrosa+2
Case study
Research

Autonomous Vehicle Collision Avoidance

Reinforcement Learning in a SUMO Traffic Simulation

A collision-avoidance policy for autonomous vehicles, trained with reinforcement learning inside a SUMO traffic simulation environment.

PythonSUMOReinforcement LearningTraffic Simulation
Case study

Skills

Technical toolkit.

Backend systems, cloud infrastructure, and the tooling to ship them reliably.

Languages

PythonTypeScriptJavaC#SQLBash

Backend & APIs

DjangoFastAPISpring BootASP.NET CoreCeleryPydanticEntity Framework

Frontend

Next.jsReactTailwind CSSFramer Motion

Cloud & Infrastructure

AWS (EC2, S3, RDS, Lambda, ECS Fargate)DockerTerraformRabbitMQPacker

Databases & Caching

PostgreSQLMySQLMongoDBRedisDynamoDB

Observability & Testing

PrometheusGrafanaOpenTelemetryCloudWatchPyTestxUnit

AI & ML

PyTorchTensorFlowBERT / TransformersHugging FaceOpenCVNLP

Tools

CursorClaude CodeGitGitHub ActionsAzure DevOpsPostman

Contact

Let's connect.

Looking for backend and cloud engineering full-time roles starting January 2027. Let's talk about how I can contribute to your team.

Email

gajera.ma@northeastern.edu

LinkedIn

Malav Gajera

GitHub

malav-250

Location

Boston, MA

Email me about Jan 2027 backend/cloud roles