Staff Backend Engineer – Real-Time Ads Recommender

Unity Technologies SF; San Francisco (CA); On-site; USD 200,000 - 261,000; Full time

Benefits offered by this job

Equity awards

Annual discretionary bonuses

Job summary

Unity Technologies SF is seeking a Staff Backend Engineer to lead Offer Delivery, shaping the architecture for real-time bidding and direct placements. You will work across auction, ML, and platform teams to implement scalable services in Go or Java, with a strong emphasis on observability and reliability.

You will mentor engineers, drive design decisions, and help evolve ranking models into production, including feature retrieval and online experiments.

Qualifications

• Proven track record of owning a backend system and delivering through others.
• Experience designing and shipping large-scale distributed services with low latency.
• Strong backend development skills in Go or Java; familiarity with gRPC.
• Experience leading across teams and mentoring engineers.

Responsibilities

• Architect & Design: set the technical direction for Offer Delivery and own architecture.
• Build & Ship: develop, test, and deploy Golang/Java services for real-time bidding and placement.
• Evolve the Recommender: collaborate with ML to productionize ranking improvements and experiments.
• Lead Across Teams: align auction, ML, platform, and product teams to drive delivery.
• Operate at Scale: ensure high availability and implement on-call and monitoring.
• Raise the Bar: mentor engineers and set engineering standards.

Skills

Technical leadership

Distributed systems

Go / Java

gRPC

Observability

Cross-team influence

Architectural design

Tools

Go

Java

gRPC

Job description

Unity Technologies SF is seeking a Staff Backend Engineer to lead Offer Delivery, shaping the architecture for real-time bidding and direct placements. You will work across auction, ML, and platform teams to implement scalable services in Go or Java, with a strong emphasis on observability and reliability.

You will mentor engineers, drive design decisions, and help evolve ranking models into production, including feature retrieval and online experiments.