Abhishek Kothari

Head of Design

United States6 mos experience

AI EnabledAI ML Practitioner

Key Highlights

Developed AI systems for construction safety compliance.
Enhanced trading strategies through predictive analytics.
Created innovative generative storytelling systems.

Stackforce AI infers this person is a Data Scientist with a strong focus on AI applications in Construction and Finance.

Contact

Skills

Core Skills

Artificial Intelligence (ai)Data ScienceFull-stack DevelopmentQuantitative Finance

Other Skills

Image Annotation ToolShinyReact.jsmulti-agent image safety pipelinevision-language modelsRetrieval-Augmented Generation (RAG)synthetic data generationmetadata extractionautomationknowledge reuseNifty Futures Trading StrategyRandom Forestdata pipelinesback-testingstrategy refinements

About

As a Data Science master’s student at Northeastern University, I focus on designing AI systems that blend technical rigor with creative and operational impact. At Suffolk Construction, I’ve built a multi-agent image safety pipeline leveraging vision-language models, retrieval-augmented generation (RAG), and synthetic data generation to identify OSHA violations from construction images, generate structured reports, and extract actionable insights from RFIs and design PDFs—helping project teams make faster, more informed decisions. Beyond my internship, I love building side projects that push the boundaries of applied AI. One of my proudest creations is MangaMatrix, a generative storytelling system that combines fine-tuned language models and vision models to convert structured narratives into panel-based manga, with automatic image generation and layout synthesis. I’ve also worked on a Visual Question Answering (VQA) system that uses ResNet and transformer-based attention to enable image-grounded NLP reasoning, as well as worked on financial forecasting models like a Nifty Futures trading strategy that improved returns through predictive analytics during my previous experience. I’m driven by curiosity, and I thrive in collaborative environments that push the boundaries of AI in industry. Outside of work, you’ll find me playing a sport (mostly cricket) or exploring investment strategies.

Experience

6 mos

Total Experience

6 mos

Average Tenure

Current Experience

Khoury college of computer sciences

Student Advisory Board Member

Sep 2025 – Present · 9 mos · Boston, Massachusetts, United States

Suffolk construction

Artificial Intelligence Intern

Jun 2025 – Dec 2025 · 6 mos · Boston, Massachusetts, United States · On-site

Built a full-stack Image Annotation Tool using Shiny and React.js, hosted on Posit Connect, streamlining custom dataset creation for internal model training workflows.
Developed a multi-agent image safety pipeline using Chatlas, integrating vision-language models and RAG to detect OSHA violations in raw jobsite images and generate automated weekly compliance reports for project stakeholders.
Designed a synthetic data generator to model realistic project metadata and GC/GR cost data, enhancing data availability and boosting downstream AI model training and scenario testing.
Collaborated with the Design AI team with “Information Extraction” initiative to extract structured metadata from Revit-generated design PDFs, supporting automation of project setup and documentation pipelines.
Contributed to Suffolk’s “Lessons Learned” initiative by embedding and retrieving insights from 700K+ RFIs, enabling knowledge reuse and improving design decision-making on new construction projects.

Image Annotation ToolShinyReact.jsmulti-agent image safety pipelinevision-language modelsRetrieval-Augmented Generation (RAG)+6

Khoury college of computer sciences

Graduate Teaching Assistant

Sep 2024 – Apr 2025 · 7 mos · Boston, Massachusetts, United States · Hybrid

Evaluated and graded assignments for 300+ students for the DS 4200 course, ensuring fair assessment and satisfaction.
Led weekly office hours, clarifying complex concepts, and boosting student performance in quizzes by 20%.
Collaborated with the professor to design quizzes, enhancing students’ grasp of key concepts by 15%.

Algoanalytics

Data Scientist

Apr 2024 – Aug 2024 · 4 mos · Pune, Maharashtra, India · Hybrid

Spearheaded the enhancement of a Nifty Futures Trading Strategy by optimizing a Random Forest classifier, boosting prediction accuracy by 5.3%.
Led the discovery and implementation of 57 key features for model training, significantly improving model performance and strategy effectiveness.
Collaborated with the CEO in refining data pipelines, reducing processing time by 8% and enhancing the overall efficiency of the trading strategy.
Conducted back-testing and strategy refinements, increasing annualized returns by 18%.

Nifty Futures Trading StrategyRandom Forestdata pipelinesback-testingstrategy refinementsQuantitative Finance+1

Pune institute of computer technology

Research Intern

Feb 2023 – May 2023 · 3 mos · Pune, Maharashtra, India · Hybrid

Pioneered a team of 3 in developing a Text-to-Image Generation Algorithm using the stable-karlo pipeline.
Accomplished an accuracy of 83% in generating relevant images based on user input.
Built a user-friendly GUI hosted on Streamlit, providing users with a custom hyper-parameter slicer.

Text-to-Image Generation Algorithmstable-karlo pipelineStreamlituser-friendly GUIArtificial Intelligence (AI)