About
This is not a chatbot, prototype, or experimentation environment.
This is a safety-sensitive, trust-dependent production system where reliability, accuracy, and controlled behavior matter.
We are looking for a senior engineer who can design and own the intelligence layer end-to-end from reasoning architecture to deployment economics.
You will work directly with the founder to shape both the technical foundation and product behavior.
This role is for builders who have shipped real systems used by real users — and understand the responsibility that comes with that.
What You'll Own
• LLM architecture and orchestration strategy
• Conversational reasoning flows with deterministic control
• Long-term memory systems (consent-aware, structured, and verifiable)
• Summarization pipelines with traceability and drift monitoring
• Latency, cost, and reliability optimization at scale
• Production evaluation, guardrails, and failure handling
• AI behavior alignment with product UX and risk boundaries
Core Responsibilities
• Design and implement production-grade LLM workflows
• Architect memory that is accurate, bounded, and auditable
• Engineer systems that minimize hallucination and maximize trust
• Build observability into model reasoning and outputs
• Optimize token usage, routing strategies, and compute cost
• Define evaluation frameworks and continuous improvement loops
• Collaborate on product decisions that affect model behavior
• Ship stable, maintainable, scalable infrastructure
Required Experience (Read Carefully)
• 3+ years building and deploying LLM systems in production
• Proven experience shipping real user-facing AI products
• Strong prompt architecture and system design capability
• Deep understanding of model limitations and failure modes
• Experience with memory design, retrieval, and summarization pipelines
• Ability to reason about tradeoffs: cost vs latency vs reliability vs safety
Strongly Preferred
• Designing AI systems that support complex business workflows
• Multi-model routing and orchestration strategies
• Evaluation frameworks and behavioral testing
• Structured knowledge storage and retrieval architectures
• Reliability engineering for AI systems
• Working closely with product leadership in early-stage environments
Do NOT apply if
• You build demos but haven't shipped production systems
• You rely on generic prompt patterns without architecture thinking
• You cannot explain technical tradeoffs clearly
• You treat LLMs as black boxes
• You are uncomfortable with accountability for real-world impact
What Success Looks Like
You design systems that behave predictably.
You reduce uncertainty, not increase it.
You build intelligence that organizations can rely on not just interact with.
If that excites you, apply with detailed, technical answers.
Contract duration of 1 to 3 months. with 30 hours per week.
Mandatory skills: Python, Machine Learning, Artificial Intelligence, API, Open AI, Amazon Web Services, TensorFlow
Languages
- English
Notice for Users
This job comes from a TieTalent partner platform. Click "Apply Now" to submit your application directly on their site.