Description :

Experience : More than 5 years of shipping production AI or machine learning systems and scaling data-intensive back ends.

Why This Role Matters :

Terrabase is building the next frontier of work AI an adaptive platform where ambient and specialized agents collaborate to deliver the one answer that matters instantly, safely, and with enterprise-grade precision. Think category-defining speed, uncompromising accuracy, and governed intelligence.

Your mission : harden that edge. Design bulletproof evaluation loops, enforce unbreakable safety nets, and squeeze every last drop of performance from our multi-agent engine.

What Youll Do :

- Build state-of-the-art deep agents and memory systems : Lead a small, high-agency team of engineers to design, train, and deploy cutting-edge agents grounded in the latest research. Read, distill, and apply SoTA papers across retrieval, reasoning, and long-term memory to push Terrabases multi-agent engine forward.

- Own the evaluation loop : Build offline and real-time test harnesses, golden-set datasets, and regression dashboards to grade each agent release on precision, recall, latency, and cost.

- Harden safety and compliance : Implement content filters, prompt firewalls, and fallback chains to ensure SOC 2 and HIPAA-grade safety.

- Optimize prompts and retrieval : Tune system, user, and tool prompts; refine ranking models and vector search parameters to lift relevance and response quality.

- Benchmark and trade off : Evaluate open-weight models, hosted APIs, and fine-tuned variants; report on accuracy latency budget trade-offs.

- Prototype and demo : Ship focused proof-of-concepts that demonstrate new capabilities before committing full sprint cycles.

- Document and scale : Write crisp design notes, post-mortems, and runbooks that make your work reproducible.

- Stay ahead : Track the latest research in RAG, tool-calling, and agent evaluation and bring whats practical into production.

What Were Looking For :

- Experience : 3 to 5 years building or operating AI / ML or data-intensive back ends in production.

- Ownership mindset : You spot issues, propose fixes, and close loops fast.

- Structured thinking : Your design docs read like reasoning in public; your code mirrors that clarity.

- Strong Python skills : Type-safe, well-tested, modern packaging you treat code as craft.

- Hands-on with frameworks : Experience with LangChain, LangGraph, or similar agent frameworks.

- Retrieval intuition : Understanding of vector databases, semantic search, and embedding pipelines.

- Evidence of depth : Prior design docs, open-source contributions, or refactored subsystems showing systematic problem-solving.

- Research literacy : Comfortable reading and re-implementing foundational work like Attention Is All You Need or its successors.

- Clear communicator : You unblock yourself and raise flags early bias for action over waiting.

Bonus Points :

- Experience with evaluation frameworks (Ragas, LM-Eval, Intercode).

- Prior work on compliance guardrails or red-team testing for Gen-AI systems.

- Contributions to open-source AI or high-signal technical writing/blogs.

Life at Terrabase :

Were a sharp, humble, fully-remote crew that values deep focus and fast feedback. Your work ships weekly to real customers supported by generous GPU budgets and a culture that prizes clarity over ceremony.

Terrabase is an equal-opportunity employer. We celebrate diversity and are committed to building an inclusive environment for every team member.