Akshay Jadhav

Software Engineer

Syracuse, New York, United States8 mos experience

AI EnabledAI ML Practitioner

Key Highlights

Developed AI-driven systems for research automation.
Engineered neural frameworks achieving 98% accuracy.
Contributed to impactful open-source projects.

Stackforce AI infers this person is a Research-focused Software Engineer with expertise in AI and Machine Learning.

Contact

akshaypjadhav@deloitte.com LinkedIn

Skills

Core Skills

AiMachine Learning

Other Skills

AI-driven systemsAWSCI/CDCNNsDockerFlaskGNNsGenerative AIHugging FaceJavaJavaScript (ES6+)JenkinsK-MeansLLaMA modelsLoRA

About

I’m a results-driven Software Engineer with 1+ years of experience building scalable web applications, backend systems, and cloud-based solutions. My expertise spans Java, Python, and JavaScript (ES6+), with strong proficiency in React.js, Redux, Spring Boot, Flask, and RESTful APIs. I’ve engineered end-to-end applications—designing responsive front-ends, architecting secure microservices, and deploying solutions on AWS with CI/CD pipelines, Docker, and Jenkins. I thrive at the intersection of cloud infrastructure, backend optimization, and modern front-end frameworks, with a proven track record of improving performance, reducing deployment cycles, and delivering maintainable, modular code in Agile/Scrum environments. Beyond professional roles, I’ve contributed to impactful open-source projects (AnkiDroid, 2M+ users), developed AI-powered research collaboration platforms, and built real-world applications supporting education, accessibility, and community-driven initiatives. I’m passionate about leveraging technology to solve complex problems, optimize workflows, and create meaningful digital experiences—while continuously learning and adapting to new challenges.

Experience

8 mos

Total Experience

8 mos

Average Tenure

Current Experience

Syracuse university

Research Assistant

Apr 2024 – Dec 2024 · 8 mos

Developed AI-driven systems for automated research team formation and NSF proposal generation using NLP, TF-IDF, embeddings, and LLaMA models.
Implemented retrieval-based and clustering algorithms (K-Means) to group researchers by expertise vectors for interdisciplinary collaboration.
Built Vision-Language Models (VLLMs) by fine-tuning Microsoft PHI-3 with LoRA, enabling domain-specific material science analysis.
Created and processed large-scale datasets of 10,000+ research papers, integrating image-text data for multimodal model training.
Utilized Generative AI (Stable Diffusion) to design bioinspired composite materials, improving impact resistance and heat dissipation in simulations.
Engineered neural frameworks combining GNNs, CNNs, and RNNs to predict structural loading curves with 98% accuracy.
Deployed models on Hugging Face and AWS, integrating scalable inference pipelines and public access endpoints.
Collaborated with interdisciplinary research teams to optimize AI pipelines, improve model interpretability, and document architectures for reproducibility.