Fellowship · Remote (India)

AI Memory Systems & Vector DB Intern

Cehpoint is hiring for this role: fellowship, Remote (India), ₹14,000 – ₹22,000/month + PPO (₹10–16 LPA). Implement embedding pipelines, chunking strategies, pgvector indexing, and semantic search retrieval benchmarks for enterprise agent memory.

About the role

Learn the core database technology powering enterprise AI. You will learn how to index millions of text chunks, optimize cosine similarity queries, and build hybrid search pipelines that combine keyword search with semantic vectors.

Work alongside senior database and AI engineers to build vector indexing pipelines, test chunking strategies, and optimize vector retrieval latency for persistent agent memory stores.

What you will do

  • Build document ingestion and chunking pipelines for PDFs, code files, and web pages.
  • Write optimized SQL and pgvector queries for nearest-neighbor semantic search.
  • Benchmark retrieval precision and recall across various embedding models and chunk sizes.
  • Assist in maintaining clean database migrations and automated backup verifications.
  • Profile query latency and measure index build times under increasing data volumes.

What you should know

  • Solid relational database fundamentals: SQL SELECT, JOIN, indexing, and primary keys.
  • Basic understanding of vector embeddings, dimensionality, and distance metrics (cosine, dot product).
  • Python programming for database connectivity and scripting.

What you should be able to do

  • Hands-on experience writing SQL queries against PostgreSQL.
  • Familiarity with Python database clients (psycopg2, asyncpg, or SQLAlchemy).
  • Ability to write clean, reusable data preprocessing scripts.

How you work

  • Interest in database internals and high-performance search systems.
  • Curiosity about the mathematical representations of text and semantic meaning.
  • Pragmatic problem solver who enjoys optimizing database performance.

Skills we check

  • PostgreSQL & pgvector
  • Python
  • Embeddings (OpenAI/Voyage/Cohere)
  • Dense Retrieval
  • SQL Queries
  • Docker Basics

What you get

  • Monthly paid fellowship (₹14,000–₹22,000/month).
  • PPO opportunity for Full-Time AI Memory Engineer.
  • Hands-on production PostgreSQL and vector database experience.
  • Full remote work support.

How we select for this role

We read every application ourselves. Here is what happens after you apply.

  1. Apply with a short form

    Paste a link to your CV and answer five questions in your own words. No sign-in and no file upload. About fifteen minutes.

  2. Integrity check

    When you press Check my answers, an automated check confirms the answers were typed by you. You see the result straight away, before you submit.

  3. Choose your project date

    If no cheating is detected you are shortlisted, and we email you a link to choose the day your project starts.

  4. 5-day practical project

    On that day our team contacts you with the brief: real work, 5 working days, reviewed by an engineer. For this role: Ingest a 500-document dataset into pgvector, create an optimized HNSW index, and implement a hybrid keyword+vector search query with sub-50ms latency (strictly 5 working days).

  5. Offer

    If your work solves the problem well, you receive your offer letter by email and in the portal.