Praveen Sharma

AI Researcher

Faridabad, Haryana, India0 mo experience

AI EnabledAI ML Practitioner

Key Highlights

Expert in building production-ready AI systems.
Hands-on experience with multilingual NLP and TTS.
Strong problem-solving skills in real-world AI applications.

Stackforce AI infers this person is a skilled AI Engineer specializing in Machine Learning and Deep Learning for production-ready systems.

Contact

Skills

Core Skills

Artificial Intelligence (ai)Machine LearningNatural Language Processing (nlp)Text-to-speech (tts) Systems

Other Skills

Model Fine-TuningSpeech TechnologyGradioEDAHugging Face TransformersMultilingual NLPAI Systems EngineeringPandas (Software)Data StorytellingPyTorchTensorFlowQuantitative AnalyticsProcedural ProgrammingComputer ScienceTableau

About

I build AI systems with a focus on real-world behavior, not just models. My work sits at the intersection of Machine Learning, Deep Learning, and system design — taking ideas from concept to working pipelines that handle messy inputs, edge cases, and practical constraints. I care about how AI systems fail, scale, and hold up in practice. Recently, I worked as an AI Research Intern at IIT Mandi, where I built an end-to-end AI pipeline and gained hands-on experience with model integration, fine-tuning, and deployment-oriented design. I’m interested in building production-ready AI products and, long-term, creating AI-driven startups where engineering depth and execution matter more than buzzwords. Open to thoughtful conversations around AI, systems, and building things from scratch.

Experience

0 mo

Total Experience

Average Tenure

Current Experience

Indian institute of technology, mandi (iksmha centre)

AI Research Intern (Speech & Multilingual NLP)

Sep 2025 – Oct 2025 · 1 mo · Mandi, Himachal Pradesh, India · On-site

Worked on designing and building a production-ready bilingual (Hindi–English) Text-to-Speech system focused on real-world code-mixed Hinglish input, where language boundaries exist at the token level.
Owned the end-to-end AI pipeline, from text processing and language identification to speech synthesis and deployment, with an emphasis on robustness and real user inputs rather than scripted examples.
Key contributions:
Designed token-level Hinglish language identification using Hing-BERT with confidence thresholds and heuristic safeguards
Built a dictionary-first transliteration pipeline with fallback mechanisms to preserve technical and domain-specific terms
Implemented regex-based text reconstruction to maintain mixed-language sentence structure
Fine-tuned and integrated Coqui XTTS v2 for natural Hindi speech synthesis with controllable prosody
Developed an end-to-end inference pipeline with a Gradio-based web interface for testing and demonstration
This work strengthened my approach to AI system design, debugging, and production readiness, particularly for India-scale multilingual and speech applications.

Artificial Intelligence (AI)Machine Learning

Skill india digital hub

Machine Learning Engineer

Mar 2025 – Jun 2025 · 3 mos · Rohini, Delhi, India · On-site

Acquired hands-on experience in data analysis, visualization, and pre-processing for diverse data types. Implemented supervised (e.g., Decision Trees, Random Forests, SVM, Linear/Logistic Regression) and unsupervised (e.g., K-Means, Hierarchical Clustering) algorithms. Explored neural networks, including backpropagation and activation functions. Conducted feature engineering and performance evaluation, addressing underfitting, overfitting, and bias-variance trade-offs. Enhanced analytical, problem-solving, and practical ML implementation skills, aligned with real-world applications.

EDAMachine Learning

Bluestock™🔺

Software Engineer Intern

Mar 2025 – Apr 2025 · 1 mo

Collaborated on the development of the IPO Web Application and REST API, delivering key features related to IPO information for the company and client sites. Participated in a team environment, following industry practices in code quality and daily reporting. Enhanced practical experience by working under a project manager’s guidance and adhering to development timelines.

Photostudy

Mathematics Specialist

Apr 2023 – Jul 2023 · 3 mos

Solved over 500+ doubts for 100+ students in live interaction sessions via video chat/text. Specialized in addressing queries related toMathematics (Calculus, Algebra, Geometry, Trigonometry, Statistics, Probability) and Physics (Kinematics, Optics). Delivered clear and concise explanations, helping students improve their understanding of complex concepts. Demonstrated strong problem-solving and communication skills in a fast-paced, virtual environment.