Senior Fullstack Software Engineer (Python) - AA, Remote: Colombia - Costa Rica,

Gorilla Logic is looking for a Senior Software Engineer, Full-Stack with strong backend expertise in Python to join an established engineering team building and scaling an automated evaluation platform for agent-oriented software products. This is a backend-heavy, hands-on engineering role focused on building production services and durable workflows that evaluate third-party applications through APIs and browser-driven interactions. You will work with significant autonomy, taking technical problems from discovery and design through implementation, testing, and production release. You’ll collaborate directly with a US-based client team and cross-functional partners while working with technologies including Python, FastAPI, durable execution frameworks, Docker, Kubernetes, Argo CD, and AWS. Responsibilities - Design, build, and maintain production backend services using Python and FastAPI. - Build and extend systems that orchestrate automated evaluations of third-party software products. - Develop long-running background jobs and durable workflows that are resumable, reliable, and observable. - Integrate with third-party APIs and systems, including authentication, rate limiting, retries, partial failures, and changing external contracts. - Build and improve browser-driven automation workflows when APIs are unavailable or insufficient. - Instrument evaluation workflows to ensure results and failures are measurable, reproducible, traceable, and easy to diagnose. - Analyze evaluation results and continuously improve configurations, fidelity, and coverage. - Design solutions that are scalable, maintainable, observable, and production ready. - Build, test, and maintain containerized applications deployed through established CI/CD and Kubernetes environments. - Use AI-assisted and agentic software engineering tools throughout planning, development, testing, and maintenance. - Take ambiguous technical requirements and independently develop practical, defensible implementation approaches. - Identify when existing technical patterns are insufficient and propose new solutions when appropriate. - Collaborate with data science, engineering, and product stakeholders to ensure implementations accurately exercise required product capabilities. - Contribute engineering perspectives to technical discovery, scoping, and prioritization. Technical Requirements - Strong professional experience building production backend applications with Python. - Hands-on experience with FastAPI or a comparable modern Python web framework. - Experience building and maintaining background job systems, workflow orchestration, or durable execution systems such as DBOS, Temporal, Airflow, Celery, Prefect, AWS Step Functions, or similar. - Strong understanding of idempotency, retries, failure recovery, and long-running stateful workflows. - Proven experience integrating with third-party APIs and external systems, including authentication and resilience to upstream failures. - Experience designing, testing, and operating production-quality distributed or backend systems. - Working knowledge of Docker and Kubernetes-based deployments. - Ability to understand and work effectively with established CI/CD pipelines. - Strong experience with testing, observability, logging, and distributed tracing. - Ability to work autonomously when requirements are ambiguous and independently drive technical discovery and implementation. - Effective use of AI coding agents and AI-assisted development tools as part of day-to-day software delivery. - Strong software engineering and system design fundamentals. - Strong spoken and written English communication skills. - Ability to collaborate directly with US-based client teams and stakeholders working in US Central Time. Nice to Have - Experience with browser automation frameworks such as Playwright, Puppeteer, or Selenium. - Experience designing reliable headed and headless browser automation and managing flaky automation workflows. - Exposure to AI or agent evaluation methodologies, including evaluation harnesses, benchmark design, LLM-as-judge approaches, or agent trajectory analysis. - Familiarity with AI evaluation frameworks such as τ²-bench (tau2-bench). - Experience working with DBOS or Temporal for durable workflow execution. - Hands-on experience with Argo CD or other GitOps-based deployment approaches. - Experience deploying applications to Kubernetes environments on AWS. - Experience developing product functionality using frontier AI models, including prompting, context management, tool use, and model reasoning. - Familiarity with the AI agent, LLM, or AI developer tooling ecosystem. - Interest in understanding, evaluating, and improving how agent-based software behaves in real-world scenarios. Originally posted on Himalayas