AI Engineer - Intern at Zycus
Zycus is a global leader in Source-to-Pay (S2P) procurement software, trusted by Fortune 1000 organizations worldwide. Powered by its proprietary Merlin AI engine, Zycus brings automation, predictive insights, and artificial intelligence across the entire procurement lifecycle.
Role Overview
Company Overview Zycus is a global leader in Source-to-Pay (S2P) procurement software, trusted by Fortune 1000 organizations worldwide. Powered by its proprietary Merlin AI engine, Zycus brings automation, predictive insights, and artificial intelligence across the entire procurement lifecycle.
Role Overview As an AI Analyst Intern, you will explore and experiment with modern AI/Gen-AI/Agentic tools across post-sales operations. Working alongside senior leaders and global delivery teams, you will build practical AI solutions, develop prototypes, and accelerate AI adoption across implementation and customer success.
Key Highlights
- Target Batch:** BE 2026 Graduates
- Work Model:** Hybrid (Bangalore)
- Domain:** AI, Automation, Generative AI, and SaaS Operations
Eligibility Criteria & Qualifications
- BE 2026 Graduates or students currently pursuing/recently completed a degree in Engineering, Computer Science, Data Science, or related fields.
- 0-1 Years of experience / Freshers.
- Basic understanding of AI/ML concepts and automation tools.
- Familiarity with Python, APIs, JSON, or data handling through academic projects or personal work.
Key Responsibilities
Model Development & Experimentation
- •Assist in identifying inefficiencies and friction points across support, implementation, and customer success workflows.
- •Create simple documentation, prompt libraries, and reports to scale successful AI initiatives.
Data Pipelines & Feature Engineering
- •Support the development of small-scale prototypes (chatbots, workflow automations, dashboards) using low-code tools or Python.
Evaluation, Deployment & Optimization
- •Help deploy and monitor pilot solutions with internal teams, tracking usage metrics and gathering feedback.
Required Skills & Technologies
Salary & Benefits
Frequently Asked Questions
Who can apply for the AI Engineer Intern position at Zycus?
BE 2026 graduates or recent graduates in Computer Science, Engineering, Data Science, or related fields with foundational knowledge of Python and AI/ML are eligible.
What is the work location for this role?
The internship is based in Bangalore under a hybrid working model.
Are frameworks like LangChain required for this internship?
Prior exposure to frameworks like LangChain, LlamaIndex, or OpenAI APIs is preferred ('Good to Have') but not strictly mandatory.
How to Apply
Submit your application directly through Zycus's verified careers portal.
Review the published role requirements, degree qualifications, and eligible graduation batches above.
Ensure your resume highlights relevant projects and contains working GitHub repository links.
Click the official application button below to open Zycus's direct careers page.
Complete and submit your application directly on the employer's portal.
You will be directed to Zycus's official careers portal. Career Insights does not process applications or charge any candidate fees.
The specifications above are verified official details published by Zycus. Candidates should verify all final terms, deadlines, and eligibility directly on Zycus's official careers portal before submitting an application.
Recommended Preparation
Independent editorial guidance based on the published role information.
What to Prepare
Evaluate your machine learning background against published criteria and modeling preparation focus areas.
B.E / B.Tech / MCA / B.Sc
Hiring evaluations prioritize demonstrable problem-solving, domain competence, and practical project evidence over specific degree pedigree.
Applied machine learning, statistical modeling, and Python libraries.
Revise ML evaluation metrics (Precision, Recall, F1, ROC-AUC), feature engineering, loss functions, and model deployment via REST endpoints (FastAPI).
Model development, dataset preprocessing, and experimentation.
Understand dataset splits, regularization techniques to prevent overfitting, vector embeddings, and modular code hygiene in ML pipelines.
Basic understanding of data storage, serialization, and persistence.
Be prepared to explain relational normalization (1NF-3NF), indexing trade-offs, foreign key constraints, and transactional consistency.
Familiarity with Git branching, pull requests, and collaborative repository management.
Review Git branch workflows, pull request code reviews, and maintain clean commit history with descriptive messages.
Onsite (Bangalore (Hybrid)) operational structure.
Highlight proactive written communication, technical documentation discipline, and the ability to troubleshoot errors systematically before escalating.
Technical Topics
Depending on the hiring team’s evaluation style, technical assessments for this role typically cover these core areas:
Topic 1: Python & Machine Learning Fundamentals
- Python data science libraries (NumPy, Pandas, PyTorch / TensorFlow)
- Supervised and unsupervised ML algorithms, loss functions, and optimizers (SGD, Adam)
- Feature normalization, embeddings, train/validation split discipline, and preventing data leakage
Topic 2: NLP, LLM APIs & Transformer Concepts
- Tokenization, embeddings, attention mechanisms, and Transformer architectures
- Prompt engineering, structured output extraction, and LLM API integration
- Retrieval-Augmented Generation (RAG) and semantic search workflows
Topic 3: Document Understanding & Computer Vision
- Document layout analysis, optical character recognition (OCR) pipelines, and bounding-box validation
- Table extraction, key-value pair parsing, and multimodal vision-language concepts
- Handling image noise, scan distortions, and irregular document structures
Topic 4: Evaluation Methodology, Metrics & Error Diagnostics
- Precision, Recall, F1-Score, ROUGE, BLEU, and exact match metrics
- Curating rigorous test sets and designing stratified validation splits
- Error analysis: categorizing model failure modes, confusion matrices, and verifiable AI
Topic 5: Experimentation, Research Documentation & Reproducibility
- Systematic experiment tracking (MLflow, Weights & Biases) and ablation study design
- Documenting research hypotheses, benchmarks, and reproducible pipelines
- Communicating model confidence thresholds, error taxonomy, and verifiable AI considerations
* Note: These are independent Career Insights editorial recommendations to help candidates structure their study plan. They do not represent guaranteed interview stages or official hiring mandates from Zycus.
Practical Project
Demonstrate the machine-learning workflow by building a small project that covers data preparation, modeling, evaluation, and reproducible results.
Document Intelligence & Agentic Table Extraction Pipeline
A reproducible document intelligence pipeline in Python that extracts structured tabular data from complex, multi-page PDF documents or scanned invoices using layout-aware models or LLM-assisted verification, complete with automated benchmark scoring.
Python, PyTorch / Transformers, Document AI & OCR, Model Evaluation & Error Diagnostics, Test-Set Curation, Reproducible Research Documentation.
Clean repository with environment setup (requirements.txt), sample evaluation dataset, benchmark scripts with precision/recall scoring, error analysis notebooks, and clear architecture documentation.
How you designed your evaluation test set, failure mode taxonomy (how the model handles rotated scans or dense tables), and the trade-offs between heuristic rules vs deep learning models.
Career Insights recommendation: This project is an independent editorial recommendation designed to help candidates build verifiable portfolio evidence. It is not required or sponsored by Zycus.
Before You Apply
Verify your model notebooks, evaluation summaries, and application materials before applying.
Ensure your resume clearly highlights hands-on projects involving Python, AI/ML Concepts, APIs & JSON. Include verified, publicly accessible GitHub repository links with clean README files.
Confirm that your graduation year (2024, 2025 & 2026 Batch), educational background, and location match the employer's published requirements before applying.
Have 1–2 key projects prepared for technical discussion. Be ready to explain your database schema choices, API contracts, error boundaries, and design trade-offs.
Submit your application directly through Zycus's official careers portal using the verified apply link below. Career Insights does not process applications or charge any candidate fees.
Career Growth Context
Career Insights editorial perspective on typical milestones and research contribution cycles in applied ML internships.
Research Orientation & Baseline Benchmarking
Familiarizing oneself with the team’s research tools, dataset pipelines, and core model baselines under technical guidance.
Hands-on Experimentation & Error Diagnostics
Conducting exploratory modeling experiments, benchmarking against baselines, and performing rigorous error analysis on complex inputs.
Validation, Reproducibility & Research Documentation
Documenting experimental methodology, optimizing inference trade-offs, and validating agentic or document intelligence workflows.
Performance-Based Conversion Opportunity
Demonstrating consistent research rigor, problem-solving, and technical ownership. As noted in available employer postings, full-time conversion represents an opportunity evaluated based on individual performance and organizational needs rather than a guaranteed progression.
This page provides verified job details sourced from Zycus's recruitment announcement, alongside independent preparation roadmaps authored by the Career Insights editorial team. Career Insights is not affiliated with Zycus and does not collect candidate data or charge application fees.
Recent & Related Jobs
Active opportunities matching similar roles, technical domains, and experience levels.
Graviton Research Capital Hiring 2026 for Software Engineer | Freshers
Freshers JobsFounded in 2014, Graviton Research Capital has grown into a global quantitative trading firm where high-performance software...
Publicis Groupe Hiring 2026 for Software Engineer | 1–5 Years
Freshers JobsPublicis Groupe is a global leader in marketing, communication, and digital transformation. As a Software Engineer in the...
Franklin Templeton Hiring 2026 for Associate Software Engineer | Freshers
Software JobsFranklin Templeton is a global leader in asset management dedicated to helping clients achieve their financial goals through...
Career Insights Guides
Curated editorial roadmaps and interview preparation guides relevant to this role.
Career RoadmapThe AI/ML Engineer's Journey: Your Essential 2026 Skill & Career Roadmap
The landscape of Artificial Intelligence and Machine Learning is evolving at an unprecedented pace, transforming industries...
Career RoadmapYour Blueprint to Success: How to Match Your Resume With a Job Description
Unlock your dream job by mastering resume-job description matching. This guide provides practical steps for freshers and...
Interview PreparationMastering the Most Asked HR Interview Questions
Ace your HR interview with this comprehensive guide. Learn to answer common questions, showcase your strengths, and avoid...