Senior ML Engineer - Real-Time Inference & Streaming (AWS)

TWG AI; New York (NY); On-site; USD 190,000 - 290,000; Full time

Benefits offered by this job

Bonus

Medical benefits

Job summary

TWG AI is seeking a Senior ML Engineer for AWS and Real-Time Inference to own the fast path: ingesting live trading data and scoring it in near real time. It's a systems-heavy role focused on streaming, low-latency inference, and the retraining cadence that keeps models current and most directly determines whether the systems keep up with live markets.

You will build streaming pipelines, low-latency inference services, and manage model retraining cadence, collaborating with data science and

Qualifications

• Experience with streaming systems (Kafka/Kinesis/MSK) and modern data storage formats.
• Proficiency in low-latency, high-throughput inference services.
• Go and Python programming; production AWS experience.
• Familiarity with financial market data a plus.
• Familiarity with chain-data infrastructure a plus.

Responsibilities

• Streaming and storage pipelines that feed both model training and low-latency inference.
• The online inference path and its latency; connect real-time detector to runtime model while keeping latency under live load.
• The model retraining cadence as data and labels accumulate, including drift-triggered retraining.
• Productionizing new features and detectors on the fast path, in partnership with data science.

Skills

Streaming systems

Low-latency inference

Go

Python

AWS

Financial data

Chain-data infra

Tools

Kafka

Kinesis

MSK

Job description

TWG AI is seeking a Senior ML Engineer for AWS and Real-Time Inference to own the fast path: ingesting live trading data and scoring it in near real time. It's a systems-heavy role focused on streaming, low-latency inference, and the retraining cadence that keeps models current and most directly determines whether the systems keep up with live markets.

You will build streaming pipelines, low-latency inference services, and manage model retraining cadence, collaborating with data science and