Staff Engineer - AI Platform

About the Role:

We're looking for a Staff Engineer (AI) to architect and build the agentic systems and AI foundations within our Platform Services team - the team responsible for the shared infrastructure, internal tooling, and core services that other product teams build on.

This is a deeply hands-on individual contributor role: you will set technical direction, write and review code, and own the design of systems that let HighLevel engineering teams build, ship, and operate AI-driven capabilities reliably at enterprise scale.

What You'll Build:

• Production-grade agentic systems - multi-step, tool-using agents with orchestration, planning, memory and context management, and human-in-the-loop controls - for platform automation and internal engineering workflows.
• A shared AI platform for product teams: reusable agent frameworks, tool and MCP integrations, an LLM gateway (routing, rate limiting, caching, fallbacks), and prompt/model versioning.
  Evaluation and observability for AI systems - eval harnesses, tracing of agent runs, and metrics for quality, latency, cost, and safety.
• AI-driven engineering practices that improve developer velocity - agentic developer workflows, AI-assisted tooling, and automation of operational toil.
• Cloud infrastructure for AI workloads across GCP/AWS, including provisioning, scaling, security, and cost governance for inference and token spend.

What You'll Drive:

Architecture & Scale: Own and evolve the architecture of agentic systems and AI platform services, ensuring they are reliable, extensible, and performant under enterprise-scale load.
AI-First Culture: Set and drive an AI-first culture within the Platform Services team - making AI tools and agentic workflows the default way the team builds and operates.
Reliability & Safety: Build agents that fail safely - with guardrails, scoped permissions, sandboxing, defenses against prompt injection, audit trails, graceful degradation, and clear incident management practices.
Evaluation & Data-Driven Decisions: Establish rigorous evaluation practices - offline evals, regression suites, and production monitoring - so agent behavior is measurable and comparable across model and prompt changes. Use that data to prioritize, evaluate trade-offs, and report on agent quality, platform health, and cost.
Cost & Efficiency: Drive continuous optimization of inference, token, and infrastructure cost without compromising quality or reliability - with measurable, sustained impact.
Technical Leadership & Mentorship: Set technical standards, lead design reviews, mentor engineers, and build a culture of ownership and craftsmanship in AI engineering across the org.
Cross-Functional Collaboration & Communication: Partner closely with product engineering teams, SRE/infra, security, and leadership to align AI platform capabilities with organizational needs, and communicate technical direction, risks, and progress clearly across the org.

What Success Looks Like:

• Agentic systems run reliably in production at growing scale, with clear quality, latency, and safety targets that are tracked and met.
• Agent changes ship with evals and observability in place, so regressions are caught before customers see them.
• Product and platform teams adopt a shared AI platform and see it as a trusted, well-documented foundation to build on.
• AI-related costs are optimized and predictable, with clear ownership of spend against usage.
• The Platform Services team works AI-first, and engineering teams measurably gain velocity from AI-driven workflows and automation.
• Engineers around you grow technically and recognize you as a go-to authority on agentic system design.

Requirements:

12+ years of overall software engineering experience, including significant time as a senior or staff-level individual contributor.
At least the last 1+ year spent architecting agentic systems in production - designing multi-step, tool-using agents, orchestration, memory and context management, and guardrails, not just integrating LLM APIs. A champion of AI agent orchestration at scale, running many agents and workflows reliably across teams.
AI-first mindset (must have) - actively using AI tools in your own engineering workflow, with the ability to set and drive an AI-first culture within the Platform Services team.
Expert-level proficiency in either Go, Python or Nodejs.
Strong hands-on orientation - comfortable writing and reviewing code, debugging production issues, and going deep on architecture. Candidates in a purely managerial (non-hands-on) role for the last 1+ year will not be considered a fit.
Strong working knowledge of LLM application fundamentals - tool/function calling, MCP, retrieval and context engineering, prompt and model management, and evaluation methodologies for non-deterministic systems.
Strong observability fundamentals, with deep experience in AI observability - tracing agent runs and monitoring quality, latency, cost, and safety in production - using tools like Prometheus, Grafana, ELK/OpenSearch, or equivalent.
Strong understanding of distributed systems architecture - consistency, availability, fault tolerance, scaling patterns, and trade-offs at scale.
Cost-aware by design - a track record of measurable cost optimization in an enterprise cloud environment, including managing LLM inference and token spend through model routing, caching, and budgets.
Practical experience with GCP and/or AWS service management - provisioning, scaling, IAM, networking, and cost governance.
Prior experience working as an individual contributor on platform teams for at least 3+ years, at enterprise scale.
Proven ability to lead technically without authority - writing design docs, driving cross-team alignment, and mentoring engineers.
A data-driven approach to decision-making, prioritization, and performance reporting.
Excellent verbal and written communication skills, with the ability to influence both technical and non-technical stakeholders.
Good to have: open-source contributions or published work on agentic systems.

Originally posted on Himalayas