Senior AI Engineer WW-FT
Arc Full-time
3 days ago
Role details
Contract type
Permanent contract Employment type
Full-time (> 32 hours) Working hours
Regular working hours Languages
English Experience level
SeniorJob location
Remote
Tech stack
JavaScript
API
Artificial Intelligence
Airflow
Amazon Web Services (AWS)
Software Design Patterns
Monitoring of Systems
Python
PostgreSQL
Redis
TensorFlow
TypeScript
Data Storage Technologies
PyTorch
Large Language Models
Kubernetes
Infrastructure Automation Frameworks
Machine Learning Operations
Docker
Job description
Join Tally as a Senior AI Engineer and contribute to an AI-first accounting platform designed for SMEs. Our mission is to automate financial workflows with extreme precision and reliability. Unlike typical LLM features, our system demands end-to-end deterministic reliability, ensuring that "almost right" is never acceptable in the financial domain., * Design and implement systems that ensure deterministic behavior in non-deterministic models, maintaining reliability over time.
- Bridge the gap from "demo to production" by developing agents capable of handling real-world complexities and edge cases.
- Prevent state drift in complex, multi-step workflows to maintain consistency.
- Convert non-deterministic LLM outputs into predictable, measurable behavior.
- Develop frameworks for evaluating AI system performance with a focus on rigor and statefulness.
- Enhance determinism and reliability within agent workflows.
- Instrument systems to ensure outcomes are measurable and metrics-driven.
Requirements
- Expertise in building stateful AI systems managing real-world workflows.
- Strong software engineering fundamentals demonstrating engineering rigor.
- Proficiency in managing agent failure modes and developing corrective loops.
- Experience with evaluations, automated benchmarking, context management, memory architectures, and advanced agent design patterns.
- A metrics-first approach to problem-solving, emphasizing that what isn't measured, isn't solved., * Familiarity with Python, TypeScript, and JavaScript.
- Experience with frameworks and tools such as PyTorch, TensorFlow, LangChain, and OpenAI/Anthropic APIs.
- Knowledge of data storage solutions like PostgreSQL, Supabase, and Redis.
- Experience with infrastructure tools like Docker, Kubernetes, and AWS/GCP.
- Familiarity with evaluation and monitoring tools including Weights & Biases, MLflow, and custom benchmarking frameworks.
- Experience with workflow and state management tools such as Airflow and Temporal.