About the Role
We are looking for a Junior AI/ML Engineer to join our team building production AI systems for an advanced, agentic travel-support chatbot.
You will work with experienced ML and software engineers on open-source language models, fine-tuning, model evaluation, inference, and production integrations. This role offers hands-on exposure to the full AI lifecycle: preparing data, training models, evaluating quality, deploying experiments, and monitoring real-world performance.
What You'll Do
- Prepare, clean, and analyze datasets from production conversations and synthetic examples.
- Support supervised fine-tuning experiments for language models and classifiers.
- Experiment with techniques such as SFT, LoRA, and QLoRA.
- Compare base and fine-tuned models using offline evaluations and production shadow traffic.
- Evaluate model quality, including accuracy, routing, tool calling, hallucinations, and reliability.
- Integrate self-hosted models into agentic workflows and LLM-powered services.
- Help configure and test models using platforms such as SageMaker, vLLM, Impala, and Baseten.
- Analyze latency, throughput, token usage, failures, and cost.
- Build monitoring, logging, and evaluation tools for model experiments.
- Write tests for model integrations, routing, fallbacks, and shadow deployments.
- Debug issues across data pipelines, model endpoints, backend services, and agent workflows.
- Document experiments, datasets, model versions, and results.
What We're Looking For
- Degree, coursework, internship, or practical project experience in Computer Science, Machine Learning, Data Science, or a related field.
- Programming experience in Python, JavaScript, or TypeScript.
- Basic understanding of machine learning concepts, including training data, validation data, overfitting, and evaluation.
- Basic understanding of LLM concepts such as prompts, tokens, context windows, structured outputs, and tool calling.
- Familiarity with REST APIs, JSON, Git, and testing.
- Strong problem-solving and debugging skills.
- Ability to analyze data and investigate model-quality issues.
- Curiosity about open-source models, fine-tuning, and production AI systems.
Nice to Have
- Experience with PyTorch or Hugging Face.
- Hands-on experience with SFT, LoRA, QLoRA, embeddings, RAG, or text classification.
- Experience preparing conversational datasets or synthetic training data.
- Familiarity with Docker, AWS, SageMaker, vLLM, or GPU-based inference.
- Experience with Jest, Grafana, New Relic, or other observability tools.
- Familiarity with Node.js, React, Redis, or asynchronous programming.
Ideal Candidate
You are an early-career engineer who combines solid software fundamentals with a strong interest in AI. You enjoy experimenting, measuring results, and understanding why models succeed or fail. You are eager to learn, comfortable working across data and code, and excited to help turn research ideas and model experiments into reliable production capabilities.
Skills Required
- Degree, coursework, internship, or practical project experience in Computer Science, Machine Learning, Data Science, or related field
- Programming experience in Python, JavaScript, or TypeScript
- Basic understanding of machine learning concepts (training/validation data, overfitting, evaluation)
- Basic understanding of LLM concepts (prompts, tokens, context windows, tool calling, structured outputs)
- Familiarity with REST APIs, JSON, Git, and testing
- Strong problem-solving and debugging skills
- Ability to analyze data and investigate model-quality issues
- Curiosity about open-source models, fine-tuning, and production AI systems
- Experience with PyTorch or Hugging Face
- Hands-on experience with SFT, LoRA, QLoRA, embeddings, RAG, or text classification
- Experience preparing conversational datasets or synthetic training data
- Familiarity with Docker, AWS, SageMaker, vLLM, or GPU-based inference
- Experience with Jest, Grafana, New Relic, or other observability tools
- Familiarity with Node.js, React, Redis, or asynchronous programming
What the Team is Saying
Navan Compensation & Benefits Highlights
-
Healthcare Strength — Employer-sponsored medical, dental, and vision coverage extend to employees and dependents, alongside mental-health resources like Headspace. Employer-verified plan listings confirm active U.S. medical coverage.
-
Leave & Time Off Breadth — Flexible vacation (unlimited PTO) is advertised in the U.S., while the U.K. page specifies five weeks of PTO, underscoring regional structure. A company-wide Quiet Week around year-end, paid holidays, and sick time are also referenced in benefits materials.
-
Parental & Family Support — Paid parental leave is clearly defined at 16 weeks for the birthing parent and 10 weeks for the non-birthing parent. Family medical leave and related supports are also noted.
Navan Insights
What We Do
Navan (Nasdaq: NAVN) is the leading all-in-one business travel, payments, and expense management platform that makes travel easy for frequent travelers. From finding flights and hotels to automating expense reconciliation, with 24/7 support along the way, Navan delivers an intuitive experience travelers love and finance teams rely on. See how Navan customers benefit and learn more at navan.com.
Why Work With Us
At Navan, we’re never satisfied with the status quo, and we know breakthrough ideas come from diverse perspectives. We are committed to cultivating a workplace that reflects the diversity of the customers we serve while fostering leadership and innovation.
Gallery
Navan Offices
Hybrid Workspace
Employees engage in a combination of remote and on-site work.
In-person connections is the foundation of Navan, the connections forged through face-to-face interactions improve company culture and what we can achieve together. We operate on a hybrid working model, which we define as four days a week in-office.
.png)
