the About Knorex
Founded in 2009, Knorex (https://knorex.com) is a technology company that provides programmatic advertising products and solutions to marketers to connect in real-time to their desired audience worldwide. Through its flagship universal marketing platform, Knorex XPO unifies the connectivity to each of the major marketing channels including social media, search, connected TVs/OTTs, video, audio, display/native, and email, in one place, while simplifying execution and optimization across these channels through AI automation. Knorex operates across the US and APAC.
Job Description
Knorex develops a cloud-based, highly scalable, all-in-one advertising and marketing platform called Knorex XPO™ (https://xpo.knorex.com/) where Advertisers, Agencies and Publishers can easily create, optimize & publish Dynamic Ads worldwide with speed and ease. As part of the platform offering, anyone can traffic their Dynamic Ads programmatically via automated buying through our in-house real-time bidder and optimizer.
Real-time bidding (RTB) is a means by which advertising inventory is bought and sold on a per-impression basis, via programmatic instantaneous auction, similar to financial markets. The entire process generally takes place in less than 100 ms. Owing to the large and real-time stream of data, coming up with efficient and pragmatic solutions and algorithms to challenging problems will become imperative. At Knorex R&D, we apply large-scale machine learning (ML) models to optimize the return-on-investment of our advertisers’ budget. Tasks we work on include user interaction prediction, bidding optimization, contextual targeting, etc.
We are seeking MLOps Engineers with the zest and passion to work on our in-house machine learning platform, which covers the whole life-cycle of machine learning, from data collection and augmentation pipeline, to model training and serving, to monitoring and analysis. You will be joining our R&D team working closely with Data Scientists, Research Scientists, and Software Engineers to build a highly scalable system that can handle billions of requests per day, all delivered in milliseconds.
In this role, you will work on large-scale and low-latency ML systems. You need to acquire a deep technical understanding of the platform, work with our cross-country team located regionally to learn about the business and technical analytics requirements, and translate them into production system.
Key Responsibilities
- Plan, design and develop components in the data pipeline to enable various machine learning models in production.
- Work closely with Data Scientists and Research Scientists to translate the state-of-the-art to production systems.
- Develop metrics to measure the outcome/impact of your introduced solutions.
- Investigate the existing pipeline, identify bottlenecks and optimize the throughput and latency of ML components.
- Document and improve the solutions over time.
- Evaluate and identify new technologies for implementation.
- Communicate with our business and technical teams to understand the analytics requirements.
- Respond and follow up to incorporate feedback and draw new insights.
- Prioritize tasks to meet multiple deadlines.
- Strong knowledge of Java is a must;
- Strong knowledge of Python is a must;
- Strong knowledge of algorithms and data structures;
- Strong in analytics and problem solving technique;
- At least 2-year working experience as ML/Data/Software Engineer or equivalent;
- Prior experience with Elastic Search is a strong plus;
- Prior experience with machine learning and NLP is a strong plus;
- Understanding of online advertising technology and RTB is a strong plus;
- Possess at least a Degree or Diploma in computer science / IT related;
- Willingness to learn and able to pick up new technology or new concepts fast;
- Able to work independently as well as in collaborative mode with minimum supervision;
- Work productively even under pressure;
- Possess good work ethic, attitude with good follow-through;
- Reasonable communication in written and spoken English.
