Prompt Engineering Intern - Internship

This is a remote full-time paid internship for an Prompt Engineer. You will

help us push the boundaries of what LLMs can do by designing, testing, and

optimizing prompts, building multi-step prompt pipelines, writing scaffolding

code around LLM calls, benchmarking outputs, and integrating AI features into

real-world products using Python and NLP techniques.

Responsibilities

* Prompt Crafting: Design, edit, and refine prompts for different use cases and models.

* Prompt Chaining: Break down complex tasks into smaller subtasks and build effective multi-step prompt workflows (e.g., summarization critique rewrite).

* Benchmarking & Evaluation: Use Python (scripts and notebooks) to run automated performance benchmarks based on KPIs like accuracy, cost, and latency.

* Feature Building: Write Python scaffolding to integrate LLM calls into usable product features and pipelines.

* Hybrid NLP: Use traditional NLP techniques (e.g., regex, spaCy, NLTK) alongside LLMs to improve output quality, preprocessing, or efficiency.

* Iterative Improvement: Run A/B tests, gather output samples, and tweak prompts or logic based on failure cases and edge conditions.

* KPI Optimization: Ensure prompt chains and model outputs meet goals like quality, relevance, length, and compute cost.

* Model Awareness: Stay updated with the latest in GPT, Claude, Gemini, and open-source LLMs.

* Tooling & Automation: Build or use lightweight tooling for prompt testing, logging, and result comparison.

* Documentation: Maintain a structured prompt and workflow logbook with evaluations, learnings, and architecture.

Salary : Rs.30,000/month

Duration : 6 months

Must-Have Qualifications

* Experience using OpenAI or other LLM APIs.

* Ability to build small-scale tools or workflows that integrate and manage prompt-based logic.

* Understanding of prompt chaining or multi-step reasoning with LLMs.

* Awareness of basic NLP techniques and when to combine them with LLM outputs.

* Comfort with debugging and improving outputs using test inputs and edge cases.

* Strong communication and analytical skills.

* Familiarity with basic evaluation techniques (BLEU, ROUGE, token count, etc.).

Nice-to-Have

* LangChain or similar framework experience.

* Experience working with vector DBs, RAG pipelines.

* Experience with Jupyter notebooks.

* Knowledge of managing context windows, formatting outputs, or chaining across models.

Ideal For

* Final-year students who have semester break for internship or recent grads ambitious about AI/LLMs and looking to develop real-world AI engineering, prompt design, and hybrid NLP-LLM skills. Youll work closely with founders to build robust, high-performing AI workflows and features for production-grade products. PPO offer post successful internship.