# Software Engineer II, Machine Learning
**Company:** [Poshmark](https://scaleengineer.com/companies/poshmark)
Software Engineer II, Machine Learning is a mid-level role at Poshmark focused on productionizing machine learning models and building infrastructure to democratize ML across the organization. This position requires 2-3 years of software engineering experience with data-intensive applications, strong understanding of ML concepts, and expertise in tools like Spark, SQL, and backend API development. The role is central to enabling Poshmark's mission to leverage machine learning across search, personalization, fraud detection, and catalog digitization for the 165-million-member fashion marketplace.
**Role:** ML Engineer
**Seniority:** Mid
**Locations:** IN Tamil Nadu (Chennai) - Office
**Salary:** 145000–185000 USD
[Apply](https://jobs.ashbyhq.com/poshmark/04687d8f-de54-4bef-a342-ec6ea6a9e29c)
Canonical: https://scaleengineer.com/jobs/poshmark/software-engineer-ii-machine-learning-04687d8f
---
## Responsibilities

- Build ML Infrastructure and Tools: Design, develop, and maintain tools and infrastructure to democratize machine learning across Poshmark's organization. Create scalable systems that enable data scientists and engineers to efficiently deploy, monitor, and manage ML models with varying requirements and complexities.
- Productionize Machine Learning Models: Collaborate with Data Science and cross-functional Engineering teams to take machine learning models from research and development into production environments. Implement best practices for model versioning, testing, validation, and deployment pipelines to ensure reliability and performance.
- Platform Maintenance and Evolution: Maintain, monitor, and support existing ML platforms and systems. Continuously evaluate and adopt newer technology stacks, architectures, and frameworks to improve performance, scalability, and developer experience across the ML engineering ecosystem.
- Design Backend APIs for ML Services: Engineer robust, well-documented backend APIs that expose machine learning models and services to internal and external consumers. Ensure APIs meet performance requirements, handle scale, and maintain data integrity across the platform.
- Optimize Data Pipelines: Build and optimize data pipelines using Spark, SQL, and streaming technologies to support machine learning workflows. Ensure efficient data processing, transformation, and feature engineering for diverse ML use cases including search, personalization, and fraud detection.

## Requirements

### education

- {"name":"Computer Science or Related Field","description":"Bachelor's degree in Computer Science, Mathematics, Physics, or a related technical field. Equivalent professional experience with demonstrated expertise in software engineering and machine learning systems is highly valued."}

### technical

- {"name":"Data-Intensive Application Development","description":"Proven experience building software systems that process, analyze, and serve data at scale. Strong foundation in handling large datasets, managing data pipelines, and optimizing performance in data-heavy environments."}
- {"name":"Machine Learning Concepts and Lifecycle","description":"Solid understanding of ML fundamentals including supervised and unsupervised learning, model evaluation metrics, cross-validation, hyperparameter tuning, and the complete ML lifecycle from problem definition through monitoring. Ability to collaborate effectively with data scientists and understand model requirements."}
- {"name":"Apache Spark and SQL Proficiency","description":"Strong hands-on experience with Apache Spark for distributed data processing and SQL for data querying and manipulation. Ability to write optimized queries, design efficient Spark jobs, and troubleshoot performance issues in data processing pipelines."}
- {"name":"Backend API Development","description":"Experience designing and developing RESTful or GraphQL APIs, understanding of service-oriented architecture, and knowledge of API best practices including authentication, rate limiting, error handling, and documentation."}
- {"name":"Python Programming","description":"Advanced proficiency in Python for machine learning, data processing, and backend development. Experience with writing clean, maintainable code following software engineering best practices."}
- {"name":"Containerization and Orchestration","description":"Hands-on experience with Docker for containerizing applications and Kubernetes for orchestrating and managing containerized services in production environments."}

### experience

- {"name":"Software Engineering with Data Systems","description":"2-3 years of professional software engineering experience specifically working with data-intensive applications, ML platforms, or analytics systems. Demonstrated ability to ship production code, handle real-world data challenges, and collaborate with cross-functional teams."}
- {"name":"ML Model Productionization","description":"Prior experience taking machine learning models to production, understanding deployment challenges, and implementing solutions for model serving, monitoring, and versioning."}
- {"name":"Collaboration in ML Ecosystems","description":"Experience working alongside data scientists and ML practitioners, understanding their workflow requirements, and building tools that enhance their productivity and model deployment capabilities."}

## Skills

### required

- {"name":"Python","description":"Expert-level proficiency in Python for ML engineering, data processing, and backend development with understanding of performance optimization and software engineering best practices."}
- {"name":"Apache Spark","description":"Strong hands-on experience with Spark for distributed computing, data processing at scale, and understanding of RDD, DataFrame, and SQL APIs."}
- {"name":"SQL","description":"Proficiency in writing complex SQL queries, optimizing database performance, and understanding data modeling concepts for analytics and ML workflows."}
- {"name":"Backend API Development","description":"Experience building production-grade APIs using frameworks like Flask, FastAPI, or Django with understanding of REST principles, error handling, and API security."}
- {"name":"Machine Learning Fundamentals","description":"Strong grasp of ML concepts including model training, evaluation, validation strategies, and understanding of common ML algorithms and their application scenarios."}
- {"name":"Docker and Kubernetes","description":"Practical experience containerizing applications with Docker and deploying to Kubernetes clusters for production workloads."}
- {"name":"Data Engineering","description":"Understanding of data pipelines, ETL processes, data modeling, and best practices for data quality and reliability in production systems."}

### preferred

- {"name":"MLflow","description":"Experience with MLflow for model tracking, experiment management, versioning, and deployment as part of the ML lifecycle. Understanding of model registry and production-ready ML practices."}
- {"name":"AWS SageMaker","description":"Familiarity with AWS SageMaker for model training, hosting, and pipeline orchestration. Experience with AWS services for ML workloads is valuable."}
- {"name":"Message Queuing Systems","description":"Experience with Redis, RabbitMQ, or Kafka for building asynchronous systems, event streaming, and managing distributed data pipelines."}
- {"name":"Deep Learning Frameworks","description":"Prior experience with PyTorch or TensorFlow for building neural networks and understanding deep learning concepts and best practices for production deployment."}
- {"name":"Stream Processing","description":"Experience with Kafka, Kinesis, or similar stream processing technologies for building real-time data pipelines and low-latency ML inference systems."}
- {"name":"Workflow Orchestration","description":"Familiarity with Apache Airflow for scheduling, monitoring, and orchestrating complex data and ML workflows at scale."}
- {"name":"NoSQL Databases","description":"Experience with MongoDB or similar NoSQL databases for handling unstructured data and building flexible data models for ML systems."}
- {"name":"Cloud Data Warehouses","description":"Experience with AWS Redshift or similar cloud data warehouses for large-scale analytics and ML feature engineering."}
- {"name":"Monitoring and Observability","description":"Familiarity with tools like Kibana, DataDog, or Prometheus for monitoring ML models, tracking performance metrics, and implementing alerting for production systems."}
- {"name":"Technology Agility","description":"Demonstrated ability to quickly learn and adopt new technologies, frameworks, and tools as business requirements evolve. Open mindset toward experimentation with emerging ML and data engineering solutions."}

## Tech stack

### tools

- {"name":"Docker","description":"Containerization platform for packaging ML models, services, and applications for consistent deployment across environments."}
- {"name":"Kubernetes","description":"Container orchestration platform for managing, scaling, and deploying containerized ML services in production."}
- {"name":"MLflow","description":"MLOps platform for tracking experiments, managing model versions, and operationalizing machine learning workflows."}
- {"name":"AWS SageMaker","description":"Fully managed service for building, training, and deploying machine learning models at scale on AWS infrastructure."}
- {"name":"Apache Airflow","description":"Workflow orchestration tool for scheduling, monitoring, and managing complex data pipelines and ML training workflows."}
- {"name":"Kibana","description":"Visualization and monitoring tool for observing model performance, system health, and debugging production ML systems."}

### others

- {"name":"Kafka","description":"Distributed event streaming platform for building real-time data pipelines and enabling low-latency ML inference systems."}
- {"name":"RabbitMQ","description":"Message broker for asynchronous processing, enabling reliable communication between ML services and other system components."}
- {"name":"AWS Kinesis","description":"Managed streaming service for collecting and processing real-time data streams to power continuous ML model predictions and feature engineering."}

### databases

- {"name":"Redis","description":"In-memory data store used for caching, session management, and enabling low-latency access to frequently used ML model features and predictions."}
- {"name":"Redshift","description":"Cloud-based data warehouse used for storing large datasets and executing complex queries for analytics and ML feature generation."}
- {"name":"MongoDB","description":"NoSQL database for flexible schema storage of unstructured data, model metadata, and configuration management."}

### languages

- {"name":"Python","description":"Primary programming language for ML engineering, data processing, and backend development at Poshmark. Used extensively for building ML models, data pipelines, and service APIs."}
- {"name":"SQL","description":"Essential for querying data warehouses, writing data transformation logic, and extracting features for machine learning models."}

### frameworks

- {"name":"Flask","description":"Lightweight web framework used for building backend APIs and ML model serving endpoints at Poshmark."}
- {"name":"PyTorch","description":"Deep learning framework utilized for building neural networks and advanced ML models with production deployment capabilities."}
- {"name":"TensorFlow","description":"Comprehensive open-source platform for machine learning used for building, training, and deploying scalable ML models."}
- {"name":"scikit-learn","description":"Machine learning library for traditional ML algorithms, preprocessing, model evaluation, and feature engineering."}
- {"name":"Apache Spark","description":"Distributed computing framework for processing large-scale datasets, feature engineering, and executing ML workflows at scale."}

## Benefits

### benefits

- {"name":"Professional Development and Learning","description":"Access to cutting-edge machine learning tools, frameworks, and technologies. Opportunity to work on complex ML challenges at scale and collaborate with leading data scientists and engineers in the fashion tech industry."}
- {"name":"Collaborative Innovation Environment","description":"Work on mission-critical ML infrastructure that powers Poshmark's 165-million-member marketplace. Influence how machine learning drives value across search, personalization, fraud detection, and catalog digitization."}
- {"name":"Growth and Impact","description":"Position as a central player in the ML organization with direct influence on technical decisions and strategic direction. Opportunity to democratize ML across the company and mentor junior engineers."}
- {"name":"Industry-Leading Problem Space","description":"Tackle fascinating ML challenges in e-commerce, social commerce, and fashion tech. Work on problems that impact millions of users and shape the future of online shopping and sustainability through secondhand fashion."}

## Compensation

- **max:** 185000
- **min:** 145000
- **currency:** USD
- **stockOptions:** true

## Interview process

### steps

- {"name":"Initial Screening Call","description":"30-minute conversation with a recruiter to discuss your background, interest in the ML Engineer II role, and career goals. This is an opportunity to learn about Poshmark's culture and the Machine Learning team's mission."}
- {"name":"Technical Phone Interview","description":"45-60 minute technical discussion with a member of the ML Engineering team covering system design, data processing concepts, ML deployment challenges, and your experience productionizing models. Expect questions about Spark, SQL, Python, and MLOps best practices."}
- {"name":"Take-Home Technical Assignment","description":"Practical coding assignment focused on ML engineering challenges. Likely scenarios include data pipeline optimization, API design for ML model serving, or ML system architecture. Usually allocated 3-4 hours with flexibility on completion time."}
- {"name":"Onsite or Virtual Technical Rounds","description":"Multiple technical interviews (typically 2-3 sessions) with different team members covering system design, ML concepts, software engineering practices, and data infrastructure. These sessions dig deeper into your technical depth and problem-solving approach."}
- {"name":"Team and Culture Fit Discussion","description":"Conversation with team leadership or senior engineers to discuss collaboration style, learning approach, and alignment with Poshmark's values. Opportunity to ask detailed questions about team dynamics, technical direction, and growth opportunities."}

## Full description
## About Poshmark

  
**Poshmark is the leading fashion marketplace where style comes alive through discovery, self-expression, and human connection.** Powered by a vibrant community of 165 million members, Poshmark brings real people and taste to shopping through a social experience shaped by shared discovery. Buying and selling fashion feels simple, joyful, and personal, while every item tells its own story. Poshmark empowers sellers to grow meaningful businesses, keeps fashion in circulation longer, and gives shoppers access to unique and trusted finds, from everyday pieces to one-of-a-kind vintage and luxury.

Our AI / Machine Learning team is a central player in the Poshmark organization. Our mission is to build a world-class machine learning platform to bring value out of data for us and for our customers. Our goal is to democratize data science and machine learning, support exploding business, and use machine learning to drive value across the chain (Search, personalization, fraud detection, catalog digitization to name a few).

The Machine Learning Engineering team at Poshmark is looking for an experienced machine learning engineer to take care of Poshmark’s requirement to take machine learning models with varying requirements to production ..

## **Responsibilities:**

* Build tools and infrastructure to democratize ML
* Productionizing ML models in collaboration with the Data Science and other Engineering teams.
* Maintain and support existing platforms and evolve to newer technology stacks and architectures.

## **Desired Skills & Experience:**

* 2-3 years of relevant software engineering experience with data intensive applications
* Good understanding of data science concepts and machine learning lifecycle
* Good understanding of spark and sql and prior experience with writing backend apis
* Flexible and open to trying out newer technologies and adopting them as and when needed.

## **Technologies we use:**

* Flask, Docker, Kubernetes
* Redis, Redshift, MongoDB, Kafka, RabbitMQ, Kinesis
* sklearn, pyTorch, Tensorflow, Spark
* Mlflow, Sagemaker, Kibana, Airflow

## **About Poshmark**

  
**Poshmark is the leading fashion marketplace where style comes alive through discovery, self-expression, and human connection.** Powered by a vibrant community of 165 million members, Poshmark brings real people and taste to shopping through a social experience shaped by shared discovery. Buying and selling fashion feels simple, joyful, and personal, while every item tells its own story. Poshmark empowers sellers to grow meaningful businesses, keeps fashion in circulation longer, and gives shoppers access to unique and trusted finds, from everyday pieces to one-of-a-kind vintage and luxury.