SDE at Calibraint | From fine-tuned LLMs to multi-agent orchestration - building AI that works

Chennai, Tamil Nadu, India

Brooklyn, New York, United States

New York, United States

Chennai, Tamil Nadu, India





Master of Science - MS, Computer and Information Sciences
New York Institute of Technology

Bachelor of Technology, Computer Science
SRM IST Chennai
Join Mageswaran and 10k+ other professionals on
Tal is the app where bosses meet engineers & builders directly, without middlemen
Download Tal BossSDE at Calibraint | From fine-tuned LLMs to multi-agent orchestration - building AI that works
At Calibraint, I design and deploy multi-agent orchestrations, build LLM-powered data analysis tools and AI applications, and lead R&D on scaling AI across the company. I work end-to-end—from prototyping agent flows with CrewAI, powered by fine-tuned, quantized local LLMs, to integrating RAG pipelines, vector databases, and cloud deployments. My focus? Creating scalable AI workflows that drive measurable impact and fuel innovation.
Chennai
gemini
langchain
langgraph
openai
Software Development Engineer 1 at Calibraint, Feb 2024 - Present · 2 yrs 3 mos, Chennai, Tamil Nadu, India, • Architect and deploy AI agents with multi‑agent orchestration and tool calling for complex workflows. • Lead R&D on next‑gen Generative AI apps, leveraging fine‑tuned LLMs and Langfuse for end‑to‑end observability. • Build, train, and integrate ML models into production-ensuring data quality, scalability, and seamless system integration. • Own the full model lifecycle-data preparation, algorithm tuning, deployment, and continuous improvement. • Deliver custom AI‑driven software solutions, collaborating with clients to turn requirements into impactful, enterprise‑ready products.
Software Intern at SOOT, Sep 2022 - Dec 2022 · 4 mos, Brooklyn, New York, United States, •Software intern specializing in applied AI/ML at an early-stage startup. Worked directly with the CTO to develop and introduce a new ML model that powers the core product. •Developed a Parametric UMAP dimension reduction ML model using TensorFlow. Scaled and trained the model on over 800 million embeddings – leveraged internal embedding data and Open AI's CLIP dataset for model training. •Trained a 3-layer fully connected neural network to reduce the training dataset from 512 dimensions to 2D space.
Graduate Assistant at New York Institute of Technology, Jan 2022 - May 2022 · 5 mos, New York, United States, Teaching Assistant for 'Topics in Computer Security', 'Cryptography'. Assisting professor and the department with various administrative tasks. Converted numerous PDFs to LaTex files. Administered Midterms and Final exams. Assisting their professor with grading papers.
Database Assistant at ONGC, Feb 2018 - Mar 2019 · 1 yr 2 mos, Chennai, Tamil Nadu, India, Assisted the design, development, and deployment of the company database. Administered, supported, and monitored databases by proactively resolving database issues and maintaining servers
Master of Science - MS, Computer and Information Sciences, New York Institute of Technology
Bachelor of Technology, Computer Science, SRM IST Chennai