Independent AI Architect
AVIC Labs LLC
- Consulting, advising, and executing complex AI, LLM, and computer vision projects for clients that need practical architecture and hands-on delivery.
Independent AI Architect
Animikh is the owner and primary service provider.
When a client works with AVIC Labs, they are working with Animikh Aich directly. He provides AI systems architecture, feasibility review, implementation, deployment, ML infrastructure optimization, and technical advisory support.
His strongest edge is computer vision and production ML, but the broader consulting offer is production AI architecture: choosing what to build, designing the system, deploying it cleanly, and making sure it is useful after the demo.
Animikh earned an MS in Artificial Intelligence from Boston University and has shipped systems across wildlife technology, video intelligence, autonomous systems research, enterprise SaaS, and AI tooling.
Experience
The work behind the consulting practice
Computer Vision & ML Engineer
Moultrie - An EBSCO Company
- Reduced infrastructure costs by $1M+ by architecting a 6x faster ML pipeline with NVIDIA Triton.
- Built animal re-identification and night-image enhancement capabilities for wildlife imagery.
- Improved animal detection accuracy and simplified production monitoring infrastructure.
Graduate Research Assistant
H2X Lab, Boston University
- Designed an autonomous driving evaluation metric with +13% improvement in ground-truth correlation, published at IROS 2025.
- Used foundation models to bridge the Sim2Real gap between CARLA simulations and real-world environments.
Machine Learning Engineer Contractor
Moultrie - An EBSCO Company
- Delivered 4x resolution enhancement on wildlife imagery.
- Developed antler segmentation and counting using Grounding DINO and Segment Anything.
Graduate Research Assistant
BIT Lab, Boston University
- Built multi-modal algorithms for causal inference research on human creativity and generative AI.
- Developed ViT/DINOv2 models for AI-generated art detection.
Computer Vision Engineer & Lead
Wobot.ai
- Led a team of 14 engineers to scale 90+ real-time video analytics use cases across 20,000+ CCTV cameras.
- Architected reusable computer vision tooling that cut development time by 50%.
- Improved alert precision to 95% and reduced false positives in production.
Co-Founder & Lead Mentor
Team Technoids
- Founded an AI/ML student club and mentored 180+ students across Python, machine learning, and computer vision workshops.
Technical Arsenal
AVIC Labs works where AI meets real system constraints: data quality, model behavior, evaluation, latency, cost, cloud infrastructure, deployment, and handoff.
PythonPyTorchOpenCVNVIDIA TritonDockerKubernetesAWSAzureFastAPILLM/RAGMLOpsComputer VisionModel ServingEvaluationAutomation
Research and Publications
Scalable Offline Metrics for Autonomous Driving
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2025)
Towards Closing the Generalization Gap in Autonomous Driving
Boston University MS Thesis, 2024
Sentiment Analysis of Restaurant Reviews Using Machine Learning Techniques
Best Paper Award, ICERECT 2019
Encoding Web-based Data for Efficient Storage in Machine Learning Applications
International Conference on Information Processing, 2019
Community and Service
Animikh co-organizes the Boston Computer Vision AIR meetup, connecting computer vision, autonomy, robotics, and AI practitioners across the greater Boston ecosystem.
He also reviews technical AI and ML systems material for publishers, conferences, and journals, keeping his consulting practice connected to current production and research patterns.
Awards and Recognition
- Best Paper Award for early machine learning research.
- First prize and winning hackathon team recognition for product and computer vision work.
- Best outgoing student recognition for academics and research.
Start With the System You Need
Discovery, architecture, build, and handoff