AI/ML Engineer (LLM Agents / FinOps)

About ByChat FinOps
ByChat FinOps is a production-ready B2B AI platform that helps organizations understand, analyze, and optimize cloud infrastructure costs across AWS, Azure, Google Cloud and Alibaba Cloud.

Our platform combines cloud billing data with Generative AI, enabling users to ask natural language questions such as:
Why did our cloud costs increase this month?
Which services generate the highest expenses?
Will we exceed our budget next quarter?
Where can we reduce infrastructure costs?

Behind the scenes, intelligent LLM Agents work with cloud APIs, read-only SQL queries, billing datasets, forecasting models, anomaly detection algorithms and FinOps analytics to deliver accurate, explainable business insights.
This is not an experimental PoC. ByChat FinOps is an actively growing production product where AI reliability, security, scalability and measurable business value are our highest priorities.

What You'll Do
As an AI/ML Engineer, you will design and build the intelligence behind ByChat FinOps by developing reliable LLM-powered agents capable of transforming complex cloud financial data into actionable recommendations.
Your responsibilities will include:
Design and develop LLM Agents, sub-agents and intelligent tools.
Build agentic workflows using prompting, tool calling and structured outputs.
Improve response quality, reliability, latency and LLM cost efficiency.
Develop secure Text-to-SQL capabilities and safe interaction with financial datasets.
Implement AI features for:Cloud cost analysis
Cost attribution
Anomaly detection
Budget forecasting
Savings recommendations

Process and analyze large-scale cloud billing datasets.
Build evaluation pipelines, golden datasets and automated LLM testing.
Improve AI observability using tracing and evaluation frameworks.
Integrate AI services with FastAPI, PostgreSQL and cloud provider APIs.
Research emerging AI technologies and successfully bring them into production.
Collaborate closely with Backend, Frontend and Product teams to deliver end-to-end product functionality.

Requirements
Must Have
Strong Python development skills.
Commercial experience building production AI/LLM applications.
Practical experience with:Prompt Engineering
Function / Tool Calling
AI Agents
Structured Outputs

Understanding of:Context management
Hallucination mitigation
Guardrails
Validation
Retry strategies

Strong SQL skills and experience with PostgreSQL or other relational databases.
Experience building APIs and integrating external services.
Understanding of ML evaluation methodologies, statistics and anomaly detection fundamentals.
Ability to write clean, maintainable and well-tested production code.
English sufficient for technical documentation and daily communication.

Nice to Have
LangChain, LangGraph or other Agent Frameworks.
AWS Bedrock, Claude or other managed LLM platforms.
Experience with FinOps and cloud billing:AWS CUR
AWS Cost Explorer
Azure Cost Management
Google Cloud Billing

Experience with:Langfuse
LangSmith
RAGAS
AI observability platforms

Experience processing large datasets (CSV, XLSX, Parquet, ETL pipelines).
Knowledge of forecasting and time-series analysis.
Docker, CI/CD and AWS infrastructure.
Experience optimizing LLM latency, token usage and operational cost.
Multi-tenant application architecture and secure AI tool execution.

Technology Stack
Python 3.12
FastAPI
LangChain
LangGraph
AWS Bedrock
PostgreSQL
SQL
Langfuse
RAGAS
Docker
Next.js / TypeScript

Why Join Us?
Build a production AI platform used to solve real business challenges.
Work with modern Generative AI and Agentic AI technologies.
Design intelligent systems that interact with cloud infrastructure at scale.
Collaborate with experienced engineers in AI, Cloud and Backend development.
Influence product architecture and technical direction.
Work on problems involving cloud economics, AI reliability, observability and optimization.
Grow together with a fast-evolving international AI product at the intersection of Generative AI, Cloud and FinOps.