Model Validation Analyst Intern

About Taabi

Taabi Mobility is a technology-driven mobility and logistics company backed by CEAT Limited, one of the flagship companies of the RPG Group. Founded in 2022, Taabi combines AI, IoT, data analytics, and connected mobility to build intelligent solutions for the transportation and logistics industry.

Taabis product portfolio spans Fuel Management, Predictive Vehicle Health, AI-powered ADAS & DMS Video Telematics, and Transport Management Systems (TMS), helping businesses improve fleet efficiency, safety, visibility, and operational performance.

As part of the CEAT and RPG Group ecosystem, Taabi brings together the credibility and industry expertise of an established business group with the agility, innovation, and technology-first approach of a growing mobility-tech company.

ABOUT THE ROLE

We are looking for a motivated and detail-oriented Model Validation Analyst Intern to join our Data Science & AI team. In this role you will validate time-series algorithms powering our Fuel Management System (FMS) covering OBD (On-Board Diagnostics) and Fuel Sensor (FS) data streams and verify event-detection outputs from our Computer Vision models. You will participate in R&D initiatives, build Proof-of-Concept prototypes, and support field testing of hardware devices and new use cases. Additionally, you will contribute to video annotation pipelines, support daily data operations, and assist in the development of AI-powered chatbot and Generative AI solutions.

KEY RESPONSIBILITIES

Time-Series Algorithm Validation (FMS OBD & Fuel Sensor)

• Validate and verify time-series algorithms used in the Fleet Management System for OBD and Fuel Sensor data pipelines.
• Perform data quality checks, anomaly detection review, and edge-case analysis on sensor telemetry streams.
• Document validation results, flag discrepancies, and collaborate with the engineering team to resolve issues.
• Create and maintain validation test cases, scripts, and regression suites for continuous monitoring.

Computer Vision Events Validation

• Validate event-detection outputs from computer vision models (e.g., driver drowsiness, harsh braking, lane departure).
• Compare model predictions against ground-truth labels; compute and report precision, recall, and F1 metrics.
• Identify false positives/negatives and provide structured feedback to the model-training team.
• Assist in building and curating event validation datasets.

Video Annotation for Data Science

• Annotate video data (bounding boxes, segmentation masks, temporal labels) following established annotation guidelines.
• Ensure annotation quality and consistency; participate in inter-annotator agreement reviews.
• Coordinate with the Data Science team to prioritize annotation backlogs and meet sprint deadlines.

Daily Operations Support

• Support day-to-day data operations including data extraction, transformation, and reporting.
• Monitor automated validation dashboards and escalate failures or anomalies promptly.
• Maintain documentation for SOPs, runbooks, and operational playbooks.

AI Chatbot & Generative AI

• Assist in building, testing, and improving AI chatbot flows and conversational experiences.
• Support prompt engineering, response evaluation, and fine-tuning experiments for Generative AI applications.
• Conduct user-acceptance testing and log feedback for iterative product improvement.

R&D, Proof of Concept (PoC) & Field Testing

• Participate in Research & Development efforts to explore new algorithms, sensor integrations, and AI-driven use cases.
• Build and validate Proof-of-Concept prototypes to evaluate feasibility of new features and hardware configurations.
• Support field testing of hardware devices (cameras, OBD dongles, fuel sensors, IoT modules) under real-world operating conditions.
• Collect, organize, and analyze field test data; document test observations, failure modes, and performance benchmarks.
• Collaborate with hardware and product teams to translate field insights into actionable improvements and new use-case definitions.

REQUIRED QUALIFICATIONS

• Currently pursuing a Bachelor's or Master's degree in Computer Science, Data Science, Statistics, Electronics, or a related field.
• Foundational understanding of time-series analysis, signal processing, or statistical methods.
• Familiarity with Python (pandas, NumPy, scikit-learn) and SQL for data manipulation and analysis.
• Basic knowledge of computer vision concepts (object detection, image classification, video analytics).
• Exposure to annotation tools (e.g., CVAT, Labelbox, V7) is a plus.
• Strong attention to detail and analytical mindset.
• Good written and verbal communication skills.

PREFERRED QUALIFICATIONS

• Exposure to LLMs, prompt engineering, RAG pipelines, or chatbot frameworks (e.g., Rasa, Dialogflow, LangChain).
• Familiarity with dashboard/BI tools such as Grafana, Metabase, or Power BI.
• Prior internship or project experience in data validation, QA, or ML model evaluation.
• Knowledge of IoT ecosystems sensors, edge devices, MQTT/HTTP protocols, and cloud-based IoT platforms.

WHAT YOU'LL GAIN

• Real-world experience validating production ML and time-series systems at scale.
• Exposure to end-to-end AI product development from data annotation to Generative AI deployment.
• Mentorship from experienced data scientists and engineers.
• Opportunity to contribute to impactful products in the fleet management and computer vision space.