Founding Agentic AI Software Engineer

Founding Agentic AI Software Engineer - Hypergrowth Social AI Startup

San Francisco, CA (you can work from home up to 3 days/week)

The company's agentic AI system facilitates real-life social connections.

The company has raised $10 million in seed funding from top VCs.

The company has 8 employees and 3 engineers and is rapidly growing; they expect to reach 20 employees in the next 6 months.

The company will pay very competitive salary. They actually are open to a salary as high as $600k, however candidates should expect that whatever salary they can get at this company would be much less cash than what they could be getting at big tech companies. The focus of the compensation package would be the equity which could be very lucrative.

Job Responsibilities:

• Ship agentic matchmaking from research to production - owning the end-to-end loop (retrieval, reasoning, tool use, safety) and driving measurable accuracy improvements.

• Building a prompt & model evaluation harness (offline + online) to compare prompts/models/policies, supporting A/B testing, and enabling fast iteration.

• Optimizing AI chat systems for lower latency, higher perceived "human-likeness," and more consistent outcomes across providers.

• Designing and maintaining context engineering pipelines (RAG, memory, summarization, compression, grounding) for conversations and matchmaking.

• Standing up observability for agents (traces, costs, failures, hallucinations, guardrails) and creating dashboards that guide product decisions.

• Collaborating daily with the cofounders and product to translate user problems into agent behaviors, experiments, and shipped features.

• Creating small internal tools and SDKs other engineers (and AIs) will use.

Qualifications:

• Self driven, self starter.

• Builder's mindset: thrives with ambiguity, ships quickly, debugs systematically, and sweats the user experience.

• Expertise with LLMs, as demonstrated clearly on your resume via professional experience or standout personal projeects.

• Experience building with multiple LLM providers and tool-calling/function-calling; comfortable swapping models and orchestrating fallbacks.

• Hands-on with RAG (indexing, chunking, embeddings, reranking) and context engineering for reliability and cost/latency trade-offs.

• Strong programming foundations (data structures, algorithms, testing, profiling).

• Practical prompt engineering and prompt libraries; can reason about failure modes and systematically improve prompts/policies.

• Ability to define metrics/KPIs (accuracy, latency, cost, safety), run A/B tests, and loop in human feedback for quality.

• Comfortable with MongoDB in production.

Nice to have:

• Familiarity with vector databases (e.g., pgvector/Redis/Pinecone/Weaviate).

• MCP (Model Context Protocol).

• Agent frameworks (LangGraph/CrewAI/Assistants).

• LLM observability/evals (e.g., Langfuse/Promptfoo/Ragas/TruLens).

• Retrieval & embeddings know-how, safety/guardrails/red-teaming.

About Skyrocket Ventures

Skyrocket Ventures is a recruiting firm for hundreds of high growth technology companies that range from industry leaders to top-tier startups. This opportunity is with one of our client companies for a full-time permanent hire. Please only apply if you are authorized to work in the U.S.

Please note that even if this job is not a perfect match, we encourage you to apply as long as it is in the ballpark. Companies are often flexible in hiring candidates who do not perfectly fit their written job description, as long as the most important qualifications are there and the candidate is good in general.

Most of the jobs we are recruiting for are not posted online, so if you would like to know of all the opportunities we have that match your interests and qualifications, then please get in touch with us.

After you apply to this job posting, we'll consider you for this job as well as any other potential matches with our client companies. If we have any potential matches, we'll share your resume with those companies and contact you about any interview opportunities we can get you.

Thank you, and we wish you a great job search