About DevRev
At DevRev, we're building the future of work with
Computer
– your AI teammate. Unlike traditional tools,
Computer
unifies all your data sources, tools, and workflows into a single AI-ready platform, giving employees real-time insights, proactive suggestions, and powerful agentic actions. It extends your existing software with AI-native apps and agents that work alongside your teams and customers – updating workflows, coordinating across teams, and eliminating repetitive work. We call this Team Intelligence: human-AI collaboration that breaks down silos, brings people back together, and frees you to solve bigger problems. Backed by Khosla Ventures and Mayfield with $150M+ raised, DevRev is trusted by global companies across industries.
Role Overview
As an Researcher Engineer Summer Intern, you will join our core AI team to explore, design, and implement state-of-the-art machine learning models and LLM-driven architectures. You will work directly on the intelligence layer powering "Computer." This role bridges the gap between bleeding-edge AI research and production-ready features that impact thousands of users globally.
Key Responsibilities
- Algorithm Development & Experimentation: Research and implement novel techniques in agentic AI workflows, hybrid search (combining keyword and semantic retrieval), and Large Language Model (LLM) reasoning and memory management.
- Knowledge Graph Integration: Develop methodologies to improve how LLMs interact with DevRev’s enterprise knowledge graphs, moving beyond traditional RAG (Retrieval-Augmented Generation) approaches.
- Data Pipelines & Evaluation: Design robust evaluation frameworks to measure accuracy, latency, and relevance for semantic indexing, reranking pipelines, and our core answering engine.
- Collaborative Research: Partner with software engineers, product managers, and senior researchers to deploy models and algorithms into real-world production environments.
- Knowledge Sharing: Summarize and present research findings, literature reviews, and prototype results to engineering and leadership teams.
Learning Goals & Objectives
- Advanced LLM Training: Gain hands-on experience training, fine-tuning, and evaluating LLMs on complex, multimodal enterprise data.
- Architectural Mastery: Understand the underlying architecture of semantic search pipelines and agentic workflows at scale.
- Research Translation: Learn how to translate academic AI research into scalable, low-latency enterprise software.
- Communication: Develop professional skills by presenting complex AI concepts to both technical and non-technical stakeholders.
Qualifications
- Education: Currently enrolled in a Ph.D. program (second year or beyond) OR a Master’s degree with 4+ years of experience in Computer Science, AI, Machine Learning, Data Science, or a related quantitative field.
Required Skills
- Programming: Strong proficiency in Python and ML frameworks (e.g., PyTorch, TensorFlow).
- AI Foundations: Solid knowledge of NLP, Deep Learning, and Information Retrieval.
- LLM Expertise: Experience with Large Language Models, prompt engineering, or vector embeddings.
- Execution: A track record of independent research or impactful academic/industry projects.
Desired Qualifications (Bonus)
- Publications in top-tier AI/ML conferences (NeurIPS, ACL, EMNLP, ICML, etc.).
- Experience with knowledge graphs, SQL-native data layers, or building autonomous AI agents.
- Active contributions to open-source AI projects.
What We Offer
- Mentorship: 1:1 pairing with a senior researcher.
- Impact: Direct influence on the capabilities of "Computer"; your work will be deployed in production.
DevRev is an equal opportunity employer and does not discriminate on the basis of race, gender, sexual orientation, gender identity/expression, national origin, disability, age, genetic information, veteran status, marital status, pregnancy or related condition, or any other basis protected by law.