freehire launches on Product Hunt on 26 August.

Follow →

ML Engineer (Intelligence) - Bengaluru

Summary

Build and improve Oolka’s core AI models and retrieval systems, focusing on LLM post-training, evaluation frameworks, and scalable retrieval/memory layers for a fintech product.

Build Oolka's core ML intelligence by leading model selection, evaluation frameworks, and the foundational retrieval/memory layer for AI systems. This is a research-driven ML Engineering role focused on first-principles problem solving.

Company Details

Oolka is an AI-powered consumer fintech company focused on helping users improve and manage their credit health using AI.

Website:

Requirements

  • 3–5 years of experience in Machine Learning or Applied AI

  • Strong ML foundations with hands-on experience in XGBoost, NLP, recommendation systems, personalization, and large-scale retrieval

  • Deep understanding of LLM architecture, training, and inference

  • Mandatory experience in LLM post-training (fine-tuning and/or reinforcement learning approaches)

  • Experience designing model evaluation frameworks and benchmarking methodologies

  • Strong knowledge of retrieval systems, embeddings, ranking, and memory architectures

  • Ability to define ambiguous ML problems from first principles and independently drive solutions

  • Strong experimentation mindset with data-driven decision making

  • Excellent communication and technical reasoning skills

Responsibilities

  • Own the model selection strategy for production AI systems

  • Design and build evaluation frameworks for LLM quality and performance

  • Architect and build Oolka's retrieval and memory layer from scratch

  • Develop scalable retrieval pipelines and ranking systems

  • Improve model quality through post-training techniques including fine-tuning and reinforcement learning

  • Drive research-backed experimentation for continuous model improvement

  • Work closely with product and engineering teams to translate ambiguous problems into robust ML solutions

  • Establish best practices for ML experimentation, evaluation, and deployment

Job Details

Location: Bengaluru

Interview Process

  • Recruiter Screening

  • Hiring Manager Discussion

  • Machine Learning Deep Dive

  • System Design & Problem Solving

  • Founder Round

Important Note

ClanX is a recruitment partner, helping Oolka hire ML Engineering.

What this application asks

recruitee

Full name, Email, CV, Cover letter, Phone

  • How many years of experience do you have as a ML Engineer working on inference? choose one
  • Are you willing to relocate to Bengaluru? yes / no
  • What is your notice period? (We are looking for candidates who can join within 30 days) choose one
  • Is your notice period negotiable? choose one
  • Do you have experience working at a product-based company? yes / no
  • Are you a citizen of India and currently based in India? yes / no
  • What is your current annual CTC range? (INR) choose one
  • Please share the figures for your current CTC breakup: Fixed, Variable, ESOPs (with vesting), and any other components
  • What is your expected annual CTC? (INR)
  • Help us with your LinkedIn profile link
  • Tell us about a project you’ve worked on that you are most proud of. Please include a brief description, your exact role, how long it took to complete, any major challenges you faced, and how you overcame them. Feel free to add a link if available. written answer
  • What interests you about this Oolka (https://oolka.in/)? Tip: Research the company and share a specific reason. Generic or copy-pasted answers will be rejected. written answer
  • Why do you want to leave your current role? written answer
  • Where did you find this job opportunity? (e.g., LinkedIn, Twitter, WhatsApp, etc). If someone referred you, please mention their name and contact number so we can thank them. (Write NA if not applicable). written answer

See also

Tailor your CV for this role?

We couldn't check your fit for this role — add a CV to your profile to see it next time.

A new version of freehire is available