Prateek Chhikara / Work Experience
Work Experience
Client-facing AI/ML work across industry use cases, research prototypes, and production systems.
Applied AI Engineer
Mistral AI- Partnering with clients across industries to translate business use cases into practical AI/ML solutions.
- Building scalable pipelines, generating synthetic data, and fine-tuning foundational models for performance, reliability, and domain adaptation.
Founding AI Engineer
Mem0.ai- Created a state-of-the-art memory solution for AI agents, enabling highly personalized interactions through adaptive context retention.
- Engineered memory-centric LLM pipeline, boosting answer quality by 26% over OpenAI and slashing token/latency costs by 90%.
- Designed scalable memory retrieval pipelines and set up evaluations to measure performance across add and search methods.
- Implemented a Neo4j-based graph-memory for efficient entity storage and retrieval, boosting context awareness and response accuracy.
AI Engineer Intern
Autoenhance.ai- Enhancing image quality globally and locally, focusing on selective detail enhancement and comprehensive filtering via deep learning.
- Developed a deep learning network for detecting similar HDR bracket images, achieving artifact-free merging.
- Engineering automated censoring for license plates, faces, and other sensitive data, ensuring image compliance with data protection laws.
Working with Prateek has been a pleasure. Working in a startup with a small team requires everyone to think on their toes and deliver quality consistently and quickly. Prateek was always keen to get involved with any challenge and communicated his ideas and results well to the team. He has a broad and in-depth understanding of computer vision and AI and is always keen to explore new and novel ways to improve the product.
Jamie McInallyCEO of Autoenhance.ai
Graduate Researcher
Information Sciences Institute- Utilized prompt engineering and memory-based approaches in large language models for text-based game playing.
- Designed reinforcement learning agents with enhanced memory, language grounding, and object affordances. (Published at KCap 2023)
- Enhanced visual question answering system by implementing visual cropping methods to focus specific region in images. (Published at NeurIPS Workshop 2023)
- Developed a multimodal approach for recipe generation from food images using attention-based vision and language models. (Published at WACV 2024)
Prateek worked with me as a research assistant for 1.5 years at USC's Information Sciences Institute, on several projects covering both interactive text-based tasks and computer vision tasks. Prateek came in already as a mature researcher, with the most solid publication history of a Master's student I've ever seen. Week by week, he was extremely respectful, hard-working, and very well-organized. His reports were always extensive with experimental results, findings, and future steps. Besides coding and discussions, he also excelled in writing - he has a clear and informative writing style, at the level of a third-year PhD student. Finally, he is a great team player, very understandable and constructive in the research process.
Filip IlievskiAssistant Professor (Sr.) at VU Amsterdam
Data Scientist
Housing.com- Achieved 92% precision on fraud engine by leveraging historical and session-level attributes to block fake leads.
- Developed a computer vision pipeline creating 3D floor plans from spherical projections, saving 400 work hours/month.
- Automated image auditing process, saving $40k/year and increasing online traffic by 12%.
- Devised an image classifier with 93% precision and 89% recall, labeling real-estate tags in real-time at 6 FPS on CPU. (Article)
- Published two research papers in top international conferences: ECML'22 and CODS-COMAD'23
I managed Prateek at Housing.com for a couple of years. His work and die hard problem solving attitude always impressed his stakeholders. Every project given to him had elements of complexity which he navigated very well. Prateek's technical skills, his ability to learn on the go and research driven attitude are going to take him a long way. He will definitely be an asset to any team he joins!
Chirag SharmaProduct Manager @ Jio PlatformsPrateek is a meticulous data scientist and brilliant researcher. He has a deep understanding of data science concepts and understands how machine learning can generate value for real-world problems. We have collaborated on multiple projects (Image classification, fraud detection, etc.) and submitted multiple research papers at top conferences. I wish him all the best and success for his future career.
Anil GoyalPrincipal Data Scientist @ Mastercard
Data Scientist
PropTiger.com- Improved image quality (contrast, brightness, sharpness, blur) resulting in 8% increase in CTR on social media campaigns. (Article)
- Developed a No-Reference Image Quality Ranking tool to select the most visually appealing real estate property images. (Article)
- Launched a service that extracts real-estate entity tags from WhatsApp group messages with 87% f1-score.