Forward Deployed AI Engineer - II

FICO; Bengaluru; Hybrid; INR 2,000,000 - 4,200,000; Full time

Job summary

FICO in Bengaluru, India, is seeking a technically oriented Customer Engineer to lead deployments, integration, and onboarding, with a focus on AI-driven automation using LLMs.

You will design agentic workflows, build lightweight tooling around LLM APIs, and mentor teammates while working in a hybrid schedule (3 days in-office per week).

Qualifications

• 4+ years in a technical implementation or customer-facing engineering role.
• Hands-on experience building with LLMs and prompts.
• Comfort with Python or similar to build lightweight tools around an LLM API.
• Fluent in enterprise software, APIs, and cloud infrastructure.

Responsibilities

• Lead customer engagements involving configuration, integration, environment setup, and solution deployment.
• Analyze engineering issues to identify AI-driven automation opportunities and build AI-powered tools.
• Design agentic workflows where LLMs reason over context, use tools, take actions, validate results, and iterate.
• Develop prompt harnesses and evaluation frameworks for real-world scenarios.
• Measure AI solutions against engineering metrics and integrate capabilities into existing tools.

Skills

LLM experience

Python

Agentic tool use

Cloud/AWS

Documentation & runbooks

Mentoring

Tools

LLM API

APIs

Automation tooling

Job description

Hybrid Work - 3 days work from office per week

What You'll Contribute

• Lead customer engagements involving configuration, integration, environment setup, and solution deployment — working from moderately-defined requirements and using sound judgment to fill in the gaps. Guide customers through onboarding, data integration, and troubleshooting, escalating only when complexity exceeds your scope.
• Analyze recurring and complex customer-impacting engineering issues to identify opportunities for AI-driven automation, and build the AI-powered tools and workflows that help engineers diagnose, troubleshoot, remediate, and prevent them — grounding solutions in real engineering context: code, logs, configurations, documentation, tickets, and incidents.
• Design agentic workflows where LLMs reason over context, use tools, take actions, validate results, and iterate, and understand and optimize the full loop: Context → Prompt → Model → Tool → Observation → Validation → Iteration. Investigate AI failures methodically, improving prompts, context, tool choice, model selection, and workflow design as needed.
• Develop prompt harnesses and evaluation frameworks that systematically test LLM solutions against real-world engineering scenarios. Build prototypes quickly, and turn the ones that work into reliable, reusable engineering capabilities — including automation that identifies patterns across customer issues and surfaces likely root causes or recommended actions.
• Measure AI solutions against meaningful engineering metrics — accuracy, resolution time, automation rate, reliability, latency, and cost — and partner with internal engineering teams to integrate proven AI capabilities into existing tools and workflows, helping establish reusable patterns for applying AI to customer engineering problems at scale.
• Build and improve reusable configuration templates, integration patterns, automation, and playbooks used across customer environments. Identify recurring implementation challenges and drive tooling, documentation, and process improvements that reduce delivery effort for the whole team.
• Develop solution guides, API examples, automation scripts, troubleshooting guides, and operational runbooks that other engineers rely on. Lead brown-bag sessions, peer reviews, and onboarding support for newer Customer Engineers.
• Take primary ownership of incident response for assigned accounts — assessing severity, driving triage, and resolving moderately complex issues yourself, escalating only the genuinely hard cases. Drive post-incident documentation and lessons-learned.

What We're Seeking

• 4+ years of experience in a technical implementation or customer-facing engineering role, with a track record of independently translating ambiguous requirements into working solutions across configuration, integration, and deployment.
• Hands-on experience building with LLMs — writing and iterating on prompts, building or using evaluation frameworks to test model outputs, and ideally some exposure to agentic/tool-use patterns (an LLM calling tools, validating its own output, iterating). Comfort working in Python (or similar) to build lightweight tools around an LLM API is expected, not just "has used ChatGPT."
• Solid working fluency with enterprise software platforms, APIs, cloud infrastructure (AWS preferred), and integrations.
• Demonstrated ability to write solutions, runbooks, and support documentation that others can rely on, and to independently identify risks before they become customer-facing problems.
• Comfortable operating with autonomy, communicating clearly with both customers and internal teams, and making sound calls under ambiguity.
• A track record of mentoring or informally guiding less experienced teammates, plus continued curiosity and professionalism under pressure.