AI Native Builder/Architect

This is a fully remote position, open to applicants in India.

📋 Description

• Lead initiatives from concept to production, overseeing the entire process from problem identification to implementation.

• Define problems, create specifications when necessary, select architectures, and develop systems.

• Collaborate with outcome owners to build agents within Atlan or at client locations.

• Define workflows, conduct early demonstrations, and deploy agents in environments where work occurs, such as Slack or Word.

• Utilize shadow runs, analyze failures, and enhance agent autonomy based on performance outcomes.

• Design mechanisms for managing failures due to unavailable tools, lack of context, uncertainties, and the need for human intervention.

• Determine when to employ deterministic code versus models, considering build versus reuse, model selection, cost implications, and reliability factors.

• Ensure agent behavior is trackable through metrics like traces, evaluations, consistency, cost, latency, and success rates.

• Expand the context layer, agent platform, and registry by incorporating versioned skills, structured instructions, permissions, secrets management, and model migrations.

• Apply coding agents to personal projects and critically assess their outputs for architectural consistency.

• Transform insights into reusable skills, reference architectures, quality benchmarks, and demonstrations.

• Operate within small two-slice teams, typically managing one initiative alongside one or two collaborators.

⛳️ Requirements

• Proven experience in developing software systems and launching 0→1 products or platforms into production, frequently without established specifications.

• Background in deploying agents or LLM systems that are utilized by actual users.

• Proficient in analyzing agent failures in relation to context, tool usage, or decision-making processes.

• Capability to utilize evaluations to identify and rectify issues while assessing user impact.

• Expertise in designing for failure scenarios, managing permissions, obtaining human approvals, and ensuring safe agent operations.

• Daily engagement with coding agents and maintaining high standards while reviewing their outputs.

• Strong reasoning based on first principles and sound systems judgment.

• Proficiency in designing for reliability in production, latency management, cost efficiency, and security.

• Ability to work independently on projects with one or two collaborators.

• Strong bonus: familiarity with MCP, A2A, agent frameworks and runtimes, LLM gateways, agent observability, agent harnesses, or orchestration layers.

• Strong bonus: experience with knowledge graphs, vector search, retrieval, or semantic layers.

• Strong bonus: background in creating evaluation suites, conducting shadow evaluations, LLM grading, or staged rollouts.

• Strong bonus: experience in deploying agents that interact with systems of record, including managing permissions, approvals, and audit trails.

• Strong bonus: experience leading AI initiatives within an engineering or business team.

• Strong bonus: experience as a founding engineer, technical founder, or involvement in early-stage startups.

• Strong bonus: a public portfolio of work through repositories, presentations, articles, or talks.

🏝️ Benefits

• Competitive base salary.

• Variable pay based on performance.

• Equity driven by impact (for most positions).

• Comprehensive health, dental, vision, and mental health benefits starting on Day 1.

• Flexible health stipends.

• Flexible time off and contemporary leave policies.

• Opportunities for accelerated growth and learning.

• A global, remote-first, high-trust, asynchronous work culture.

• The ability to work from anywhere with a diverse team spanning over 15 countries.