Senior Software Engineer - AI Products

About the role

We're hiring senior and staff engineers to join our AI product engineering team. Here's what makes this different: at Gem, software engineers don't just build around models, they work directly on them. You'll fine-tune LLMs and embedding models, rethink our search architecture, clean and optimize data flows, and make calls on what systems and infrastructure we use. This team provides high ownership with the opportunity to experiment, prototype and then ship to production. Nothing is off the table if it makes our search faster and more accurate. This is hands-on work from end to end. You'll integrate directly with LLMs and rerankers, experiment with new models as they launch, build evaluation systems to measure what actually matters, and own the entire stack from Snowflake data pipelines to embedding queries to the UI someone sees in their browser.

What You’ll Build

• Model work: fine-tuning LLMs and embedding models for recruiting queries, testing new providers as they launch, building systems to evaluate what actually improves search quality
• Search at scale: making semantic search instant across 800M+ profiles, integrating rerankers to surface better candidates, designing the feedback loops that help search get smarter
• Data infrastructure: owning pipelines in Snowflake that feed our models, cleaning and structuring candidate data, building the systems that let us experiment quickly without breaking production
• Shipping full-stack features: writing the code from prompt engineering to UI, creating interfaces that make complex search feel simple, iterating based on what recruiters actually tell us

What You Bring

• About You: You have 5+ years of industry experience as a software engineer, building user-facing products. You enjoy mentoring, working cross-functionally, and working directly with customers. You move with velocity and invest in improving your craft. You prioritize technical quality, building reliable and scalable systems. You are product-oriented and you don’t hesitate in making decisions. You are energized by collaborating in-office with your peers in a hybrid model.
• Extra Credit: You've built agents or worked on eval infrastructure, you know your way around vector databases (Pinecone, Weaviate, Qdrant) or search systems (Elasticsearch), you've worked with RAG architectures or ML observability, you've worked at an early-stage startup where you had to figure things out yourself, Background in information retrieval, ranking, NLP, or recommendation systems, Experience with data pipelines (Snowflake, dbt), model deployment, or monitoring models in production, Comfortable in Python and working with ML frameworks like PyTorch or TensorFlow

How We Work

• We've removed most development friction. Local dev with Vite boots instantly with hot-reload. CI runs in ~10 minutes, deploys go straight to production. We ship fast and iterate based on customer feedback.
• Our Engineering Culture: Reasonable hours, Weekly team events and happy hours, Regular hackathons for experimental features, Team from Meta, Uber, and Dropbox

Benefits

• 10-year stock option exercise window
• Flexible Time Off and 16 paid holidays (including company wellness days)
• Best-in-class medical, dental, and vision coverage
• $1,200 annual learning and development stipend
• 16 weeks paid parental leave for all parents, plus $1,500 new-parent perk and flexible return-to-work options

Role Details

• Location: This role is based 3 days per week out of our San Francisco HQ and is not eligible for full-time remote work.
• Career Path: The annual cash compensation range for this position is $190,000–$260,000 based on level in addition to equity & benefits. The range displayed on this job posting reflects the minimum and maximum compensation. Factors including location, level, job-related knowledge, skills, and experience will determine compensation.