//Software Engineer
Vistrit Pandey — AI engineer and software engineer building applied AI that ships to production.
I build applied AI and ML systems end to end — LLM training and evaluation pipelines, local inference infrastructure, and full-stack products that ship to real users.

Vistrit Pandey
Software Engineer
Applied AI & ML SystemsResearch & Product EngineeringForward-Deployed Engineering
Where I've worked
01 — Experience
Software Engineer II, AI R&D
Blinkfire Analytics
Jan 2026 — PresentChicago, IL
Leading applied AI research and infrastructure across LLM training, inference, and production AI services.
- Built reinforcement learning, synthetic data generation, and model evaluation workflows over 12+ years of proprietary multimodal data for LLM training, benchmarking, and fine-tuning.
- Designed local LLM inference infrastructure using CUDA, TensorRT, NVIDIA Triton, Ollama, Docker, Kubernetes, and Linux to improve serving latency, throughput, deployment control, and cost.
- Developed AI agents and retrieval systems combining RAG, semantic search, tool execution, and customer-scoped context to generate analytics, recommendations, reports, and structured files.
- Built multimodal systems using Whisper, LLMs, and vision-language models for audio understanding, contextual classification, automated review, and reduced manual processing.
- Partnered with product, infrastructure, leadership, and enterprise stakeholders to ship production AI services, backend APIs, and full-stack product workflows.

Software Engineer I, AI R&D
Blinkfire Analytics
Apr 2024 — Dec 2025Chicago, IL
Shipped AI and analytics products across the stack with real-time inference workflows.
- Shipped AI and analytics products using Python, Go, TypeScript, REST, gRPC, Redis, Next.js, Vue.js, Firebase, and Google Cloud.
- Integrated GPT, Gemini, Llama, Hugging Face, and open-source models into semantic search, automation, content intelligence, and real-time inference workflows.
- Built model fine-tuning, quantization, serving, and inference optimization workflows using PyTorch, TensorFlow, CUDA, TensorRT, Triton, Docker, and Kubernetes.
- Developed backend APIs, asynchronous processing systems, annotation interfaces, and customer-facing applications across distributed service architectures.
Course Assistant, Computer Science
Drexel University
Sept 2022 — Sept 2023Philadelphia, PA
Supported computer science coursework through mentorship, labs, and evaluations.
- Mentored 500+ students across algorithms, machine learning, computer vision, computational imaging, programming, and software engineering through labs, debugging, code reviews, and project evaluations.
Where I studied
02 — Education
Graduate
University of Illinois Urbana-Champaign
Master of Computer Science
Aug 2026 — May 2028Illinois, US
Undergraduate
Drexel University
B.S. Computer Science, Minor in Information Systems
Sept 2019 — Sept 2023Philadelphia, PA
Products I've built
03 — Selected workThings I'm exploring
04 — Research & open sourceWhat I work with
05 — ToolkitLanguages
PythonGoTypeScriptJavaScriptSQLJavaC++Bash
AI & ML
PyTorchTensorFlowHugging FaceMLXReinforcement LearningFine-tuningModel EvaluationSynthetic DataQuantizationRAGSemantic SearchMultimodal AIWhisperVision-Language Models
Systems & Product
CUDATensorRTNVIDIA TritonDockerKubernetesLinuxGCPAWSRedisRESTgRPCDistributed SystemsReactNext.jsVue.jsFirebase

