Building autonomous AI systems & high-performance data pipelines.
I'm Aditya — a Data Scientist and ML Engineer currently finishing my M.S. in Data Science at Stevens Institute of Technology in Hoboken, NJ. I got into this field because I genuinely enjoy the process of taking messy, real-world problems and turning them into something a model can reason about.
Most of my work revolves around LLM-driven autonomous systems and scalable data pipelines. I've built things like multi-agent RAG systems that orchestrate specialized LLM workers, browser extensions that have to fight through Shadow DOM and framework-level restrictions to automate form filling, and speech-to-speech translation pipelines where every millisecond of latency matters. I care a lot about writing code that actually ships — not just notebooks that run once.
Outside of the typical ML stack, I spend time thinking about AI safety and alignment. I find the challenge of getting deterministic, trustworthy behavior out of probabilistic models genuinely interesting, not just as an academic question but as something that matters the moment you put a model in front of real users. I'm also an AWS Certified AI Practitioner and ML Engineer, which keeps me grounded in how these systems run at scale in production.
| Project | Description |
|---|---|
| Termnova · live demo | Production-grade AI contract intelligence platform with hybrid RAG, LangGraph multi-agent workflows, OpenTelemetry tracing, and hallucination guardrails. Iterating fast — hybrid retrieval moving into PostgreSQL, durable idempotent ingestion, OpenCode routing with DeepSeek fallback. |
| Cadence | Evidence-grounded CI intelligence — reads your build history over the plain GitHub API, quantifies what it costs in minutes and dollars, and opens the PR that fixes it. No install, no config, every number replayed from real runs. |
| Alpha-Aware Hierarchical RL | Autonomous trading agents in ABIDES limit order book simulations studying market microstructures and regime classification. |
| Supplier Intelligence Platform | Multi-source B2B intelligence engine extracting supplier risk signals, financial health indicators, and market trends. |
| AI DOM-Autofill Extension | Browser automation engine bypassing ATS limitations via recursive Shadow DOM traversal and React synthetic event dispatcher overrides. |
Landing fixes and features in ML, data, and evaluation tooling — merging into maintenance-heavy codebases with full test discipline.
| Repo | PR | What it does |
|---|---|---|
| kornia | #4135 (merged) | Revived LoFTR's two end-to-end accuracy tests on CPU — a dead version guard had silently skipped them on every supported torch. |
| Arize AI Phoenix | #15683 (merged) | Fixed playground/evaluator clients dropping Anthropic & Bedrock tool-choice and strict settings before the request went out. |
| sqlfluff | #8420 (merged) | Added full grammar for Snowflake CREATE/ALTER/DROP ALERT DDL. |
| sqlfluff | #8411 (merged) | Fixed reflow alignment pads when an alignment predecessor sat on a different line (leading-comma T-SQL). |
| Stanford DSPy | #10275 (open) | Rejects reserved trajectory as an output field in ReAct — surfaces the collision at construction instead of mid-forward() after billed LM calls. |
| Languages |        |
| ML & AI |        |
| Frameworks |     |
| Data |     |
| Infrastructure |        |
Engineered an enterprise-grade contract analysis platform utilizing LangGraph multi-agent orchestration and Hybrid RAG (dense vector embeddings combined with sparse BM25 reranking). Implemented asynchronous ingestion queues with Celery, end-to-end distributed tracing via OpenTelemetry, and strict quantitative RAG evaluation metrics (faithfulness, context precision, hallucination scoring) to audit complex legal agreements with deterministic precision.
Won 1st place in a Databricks Hackathon by designing a Vision-Critic multi-model jury architecture for insurance claim fraud detection. Visual forensic analysis (OpenCV keyframe extraction and metadata artifact detection) was strictly decoupled from logical LLM deduction to eliminate decision contamination. Built weighted heuristic risk-scoring models that evaluate cross-modal signals across video, images, and transaction metadata.
Architected a concurrent, low-latency audio translation pipeline using a producer-consumer threaded design across Whisper (ASR), MarianMT (Neural Translation), and Meta MMS (TTS). Achieved <3ms Voice Activity Detection latency with Silero VAD, sub-3s end-to-end latency, and dynamic gain-normalization filters to eliminate background static.
Developed a distributed medical computer vision pipeline on Google Cloud Platform (GCP) leveraging PySpark and Spark ML for large-scale fundus image preprocessing. Applied PyTorch deep neural networks with custom CLAHE (Contrast Limited Adaptive Histogram Equalization) contrast-enhancement algorithms to classify diabetic retinopathy severity while preserving critical microvascular pathology.
