Senior Software Engineer (Data Search) at Modash

About Modash

Modash helps brands find, understand, and work with creators across Instagram, TikTok, and YouTube. Over 2,700 companies—including Stanley 1913, Sennheiser, and NordVPN—rely on Modash to manage and scale their creator partnerships.

The core challenge lies in solving a complex search problem: helping customers discover the right creators from 400M+ profiles and billions of media files. This involves combining large-scale data processing, traditional retrieval, vector search, multimodal embeddings, and large language models (LLMs).

The Role

Search at Modash isn’t just an internal tool or support function—it’s a core product customers rely on to discover creators. The scale is massive, the data is messy, and search intent can be highly nuanced. Customers may seek creators in specific niches, with particular visual styles, or resembling existing accounts. Solving this requires expertise across retrieval systems, embeddings, ranking, relevance, and low-latency serving.

You’ll join the specialized Data Search team, collaborating closely with Data Core, Data Insights, product teams, customers, and leadership. Your work will demand real autonomy, but you won’t operate in isolation.

Key Responsibilities

1. Enhance Creator Search: Improve how customers discover creators across 400M+ profiles and billions of media files, focusing on retrieval, filtering, ranking, relevance, speed, and product decisions.

2. Multimodal Data Integration: Build systems that generate and utilize embeddings from images, video, text, and audio at scale, ensuring these signals are actionable in customer-facing search experiences.

3. Ship Production Capabilities: Evaluate models and technologies pragmatically, balancing trade-offs between cost, latency, and quality, and transition promising approaches to production within weeks, not quarters.

4. End-to-End Ownership: Shape problems, gather requirements, design architecture, write code, release systems, measure outcomes, and iterate—owning results, not just implementation tasks.

Day-to-Day Work

Your typical week might include:

• Monday: Diagnose and improve customer search results by refining retrieval and ranking stages to better align with intent.
• Tuesday: Develop pipelines for generating multimodal embeddings, testing their impact on retrieval quality and cost.
• Wednesday: Collaborate with Data Insights to define and implement new in-house data points, exposing them in Search for enhanced creator discovery.
• Thursday: Test reranking models on real customer queries, measuring relevance improvements against latency and inference costs before deciding on production readiness.
• Friday: Review production metrics, investigate relevance regressions, and share insights with the team, addressing issues in models, data, query logic, or product design.

Meetings are kept purposeful, and deep work is protected. Expect short standups, close collaboration when needed, and ample time for designing, building, optimizing, and launching.

Requirements

Past Experience

• Built large-scale data or backend products with experience in systems where volume, latency, reliability, and cost are critical.
• Shipped products from concept to production, owning scoping, architecture, implementation, release, measurement, and iteration—not just one layer.
• Designed distributed systems, reasoning about throughput, failure modes, data flow, scalability, and operational trade-offs.
• Built LLM-powered or agentic features in production, understanding model differences and balancing capability against latency and cost.
• Worked autonomously on ambiguous problems, gathering requirements, asking insightful questions, and advancing incomplete contexts.
• Communicated technical trade-offs clearly across teams, explaining to both engineers and non-engineers, and providing direct feedback without unnecessary process.
• Thrived in fast-moving product environments, adapting quickly, shipping incrementally, and pivoting based on evidence.

Bonus points for experience with multimodal embeddings, vector databases, semantic search, ranking algorithms, model deployment, self-hosted models, or GPU infrastructure. Familiarity with the creator economy is a plus but not required.

Our Tech Stack

• Cloud: AWS and GCP, with Pulumi for infrastructure as code
• Languages: Python, TypeScript, and Node.js
• Big Data: PySpark on AWS EMR
• Workflow: Airflow
• Vector DB: Milvus (via Zilliz)
• Search: Elasticsearch
• LLMs: Batch APIs
• Data Lake: Apache Iceberg
• AWS Services: SageMaker, DynamoDB, S3, Glue, Kinesis, Lambda, ECS, and Aurora
• Tools: Slack, GitHub, Linear, Notion, and Cursor

Benefits

• Fully Remote in Europe: Work from anywhere with overlap in GMT+3.
• Compensation: Salary range of 100,000€ - 130,000€ annually, including stock options. Exact amount depends on location, employment type, skills, and experience.
• Flexible Hours: Focus on outcomes, not hours logged.
• Unlimited Paid Vacation: Take time off as needed to stay rested and productive.
• Personal Development Support: Access to courses, books, and conferences to support growth.
• Real Ownership: Solve challenging search problems from idea to production with minimal process barriers.
• Regular Offsites: In-person gatherings to connect, collaborate, and enjoy time together.

Modash was founded in 2018 by a high-school dropout and a Canadian, building tools to help brands scale partnerships with online creators. With 8-figures in ARR and a $12M Series A investment, Modash is positioned for long-term growth, aiming to remain a leader in the creator economy for decades.

The company operates with a fast, async-first culture across 20+ countries, fostering a collaborative environment where employees are driven to excel in their craft and inspire others. Join Modash to make an impact, create lasting memories, and contribute to meaningful work.