Founding AI Engineer
- Discuss this with your agent
- Open in Claude
- Open in ChatGPT
Prepare application
- Draft this with your agent
- Open in Claude
- Open in ChatGPT
Role details
Tech stack
+2 more
Job description
- Architect & Build Agentic Systems: Design, prototype, and harden production-grade orchestration frameworks that leverage LLMs, persistent memory, and tool-integration to navigate complex, non-linear enterprise tasks
- Build the Data Flywheel: Implement sophisticated RAG (Retrieval-Augmented Generation) pipelines, ensuring high-fidelity context retrieval from fragmented data sources
- Evaluate & Iterate: Build internal benchmarking and evaluation suites to measure and improve model reliability, grounding, and latency
- Strategic Modeling: Architect high-performance intelligence stacks by making pragmatic trade-offs between Frontier Reasoning Models for long-horizon agentic planning, Fine-tuned Open-Weight Models for data sovereignty and low-latency inference, and Deterministic Heuristics for reliable, cost-efficient guardrails
- Cross-Collaboration: Partner with product and engineering to turn ambiguity into working software
Requirements
- Systems Engineering: 7+ years of experience and a deep background in building production software. You understand that βAIβ is only as good as the infrastructure surrounding it
- Modern AI Fluency: Deep practical experience with the transformer-based stack: embeddings, vector databases, tool-calling, and context window management
- Production LLM Experience: 2+ years shipping LLM-based products to real users in production
- Python Power-User: You are comfortable building entire systems, not just experimenting in notebooks
- Reliability Mindset: You have strong instincts for failure modes (hallucinations, API timeouts, βinfinite loopsβ in agents) and how to build guardrails against them
- Security Mindset: You treat access to enterprise data as part of the design, not an afterthought
- Clear communicator with founder-level ownership mindset, * Experience in building complex, multi-step workflows using frameworks such as LangGraph, LlamaIndex or PydanticAI
- Hands-on experience with Fine-tuning and PEFT (LoRA/QLoRA) to adapt open-weight models for domain-specific tasks
- Experience with human-in-the-loop workflows and enterprise identity (SSO, role-based access)
- A background in Enterprise Operations (Industrial, Logistics, Fintech, or CRM workflows).
- Strong opinions on where AI creates value and where itβs just expensive noise, We value speed, clarity, and technical rigor. We favor simple systems over clever ones, prototypes over over-planning, and ownership over handoffs. Youβll work directly with founders and ship things that matter quickly.
Benefits & conditions
Compensation Range: $220K - $250K
About the company
Skiffra is building the intelligence layer for operational decision-making in physical industries.
Weβre developing AI-native orchestration systems that turn complex, high-stakes operations into intelligent, adaptive workflows. Weβre starting with mining and natural resources, where decisions move people, equipment and capital, and where the consequences of getting them wrong are real.
Our ambition extends well beyond mining. Weβre building a platform designed to change how physical industries make decisions.
As a Founding AI Engineer, you will be the architect of our core intelligence. This is a role for a high-leverage builder who can move fluidly between prompt engineering, backend systems, and model optimization. You will design the βbrainβ of Skiffra, moving agentic workflows from whiteboards to production-grade systems that handle real enterprise data at scale. This is a founding-level role for builders who value shipping over dogma and real-world impact over demos.
This is a full-time role in our Los Angeles office.
Apply for this position
This job is hosted externally. Click below to view the full posting and apply.
Prepare application
- Draft this with your agent
- Open in Claude
- Open in ChatGPT
Good distractions
Loading talks and stories from around this roleβ¦