Job Description
Ovii's Interpretation of the Role
Lead AI Engineer driving end‑to‑end implementation of human‑in‑the‑loop AI systems for Kobie's loyalty platform. Own design specs, ship core routing and execution components, and mentor engineers while ensuring safety and reliability at scale.
Role Snapshot
- Lead AI Engineer (individual contributor)
- Hybrid role based in Bengaluru
- Full‑time, senior‑level
- Focus on LLM production systems
- Mentorship & code review
Must-Have Requirements
- Python
- LLM systems operation (prompt/context engineering, tool/function calling, RAG, evaluation, observability)
- LangChain/LangGraph or comparable framework
- SQL
- Git
- Docker
- Modern API frameworks
- Strong written communication
- Python production development
- LLM systems deployment
- Designing oversight mechanisms for AI agents
Nice-to-Have Signals
- Amazon Bedrock or AgentCore runtime experience
- Snowflake / Snowpark / Snowflake Cortex
- Experience in loyalty, martech, or adtech domains
- Amazon Bedrock / AgentCore development
- Snowflake ecosystem
- Loyalty or martech domain experience
- loyalty
- martech
- adtech
Work Setup
- Location: Bengaluru, India
- Work mode: HYBRID
- Remote scope: UNSPECIFIED
- Employment type: Full-Time
Not Specified in JD
- Salary range
- Visa sponsorship
- Remote eligibility
- Education requirement
- Certifications
- Relocation
- Notice period
- Travel
- Security clearance
- Coding test
- Portfolio
- GitHub
- Writing sample
- Cover letter
What You'll Likely Work On
- Author detailed per‑feature implementation specifications for AI agents
- Build and ship the human‑in‑the‑loop routing engine and execution wrapper
- Design oversight mechanisms such as refusal policies and prompt‑injection safeguards
- Implement queue mechanics, reviewer assignment, back‑pressure and run‑resumption semantics
- Create evaluation and observability pipelines for LLM outputs
- Review pull requests and mentor engineers on complex AI implementation challenges
- Set standards for what AI functionality is shipped versus refused
Good Fit If You Have
- Enjoys tackling hard implementation problems and shipping production code
- Comfortable authoring clear technical specifications for other engineers and AI agents
- Has a pragmatic view of AI failure modes and designs safeguards accordingly
Skills
- Python (production‑grade)
- LLM systems operation & prompt engineering
- LangChain / LangGraph or comparable agent framework
- LLM observability (CloudWatch, LangSmith, Langfuse, MLflow, OpenTelemetry)
- SQL
- AWS cloud platform (preferred)
- Git & Docker
- Modern API frameworks
- Strong written communication