Finkraft is a fast-growing fintech startup building infrastructure to simplify and modernize financial operations. We are embedding production-grade AI and agentic systems into core workflows to drive automation and real business outcomes.
Role Overview
We’re looking for an AI Engineer to design and scale agentic AI systems that move beyond experimentation into real-world usage. You’ll build intelligent agents, integrate LLMs into production, and own AI systems end-to-end.
What You’ll Do
Build and deploy AI agents for multi-step workflows
Design RAG pipelines, tool usage, memory, and orchestration logic
Integrate LLMs into production systems with monitoring and safety
Define evaluation metrics, feedback loops, and guardrails
Select and manage LLMs, vector databases, and frameworks
Collaborate with product, data, and engineering teams to ship AI features
Deploy AI models on AWS infrastructure and microservices
Work with data pipelines (Python, Airflow) to prepare training data
Implement AI testing and monitoring (Selenium, Playwright for integration testing)
Collaborate with backend teams on API integration (REST APIs)
Ensure AI systems integrate with PostgreSQL & MongoDB data stores
Requirements
What We’re Looking For
4-6 years in AI Engineering
Hands-on experience with LLMs, RAG, vector stores, and agent frameworks
Strong Python programming and system design skills
Experience deploying AI systems in production
Strong problem-solving with focus on real-world outcomes
Ability to balance speed, reliability, and scalability
Good to Have
MLOps Tools (MLflow, Weights & Biases for model tracking)
Prompt Engineering (advanced techniques and optimization)
Fine-tuning LLMs (custom model adaptation)
Evaluation Frameworks (RAGAS, LangSmith for AI evaluation)
Observability Tools (monitoring AI system performance)
Playwright/Selenium (testing AI-driven workflows)
Kafka (streaming data for AI pipelines)
Fintech or enterprise workflow experience
AI safety and alignment principles
Benefits
Why Join Us
Build real AI systems used in production
Shape how AI integrates into fintech workflows
High ownership, work closely with founders
High-impact role with strong growth potential
Work on cutting-edge agentic AI systems
Skills Required
- 4-6 years in AI Engineering
- Hands-on experience with LLMs, RAG, vector stores, and agent frameworks
- Strong Python programming and system design skills
- Experience deploying AI systems in production
- Deploy AI models on AWS infrastructure and microservices
- Work with data pipelines (Python, Airflow) to prepare training data
- Implement AI testing and monitoring (Selenium, Playwright for integration testing)
- Collaborate with backend teams on API integration (REST APIs)
- Ensure AI systems integrate with PostgreSQL & MongoDB data stores
- Strong problem-solving with focus on real-world outcomes
- Ability to balance speed, reliability, and scalability
- MLOps Tools (MLflow, Weights & Biases for model tracking)
- Prompt Engineering (advanced techniques and optimization)
- Fine-tuning LLMs (custom model adaptation)
- Evaluation Frameworks (RAGAS, LangSmith for AI evaluation)
- Observability Tools (monitoring AI system performance)
- Kafka (streaming data for AI pipelines)
- Fintech or enterprise workflow experience
- AI safety and alignment principles
What We Do
We are called FinKraft. What do we do? FinKraft - Krafting reliable automated solutions to make complex business processes easier. How do we do it? The process of gathering information, collecting data from various sources and reconciling records is a very painstaking and complex process. With FinKraft, companies need not worry about any of it. We have built a robust automated SaaS platform to help our clients with these processes and eliminate human error. One of the first use cases we have mastered is GST credit on travel expense for corporate India. “Capgemini” saves one million dollars annually using our tool. This marked the beginning of our continuous innovation. We aim to make a difference across industries. A few of our other clients include: Consulting- Mckinsey, Boston Consulting Pharma- Abbott, Sun Pharma Retail- Ikea, Godrej, Benetton Government- WHO, World Bank, US Embassy Banking- American Express, Kotak Industrial- Emerson, Schneider Others- WPP Group, Hershey’s, Tata’s Our Robotic Process Automation (RPA) system makes the process of the extraction of information from emails, websites and other portals more convenient than ever before. Post extraction, our system will also help you read and extract data, reconcile and create actionable real-time dashboards. Strong archiving with an excellent document management system for quick and easy retrieval. Inconsistency and errors in the data are highlighted by our portal to help clients take appropriate measures at the right time. This saves time, money, and enables companies to get error-free reports in just a click!
