Skip to content
View sanchit056's full-sized avatar
2 followers · 0 following

Achievements

Achievement: Pull Shark

Achievements

Achievement: Pull Shark

Block or report sanchit056

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
sanchit056/README.md


  Publications

fastWorkflow: Closing the Performance Gap Between Small and Frontier Language Models for Conversational Agents

Sanchit Satija, Aditya Bhatt, Priyanshu Jani, Dhar Rawal
ACM Conference on AI and Agentic Systems (CAIS) — 2026, San Jose, CA

A dual-mode agentic framework that addresses systematic small-model failure modes in conversational workflows through a cascaded NLU pipeline, hierarchical context organization, dependency-aware planning, and adaptive context management.

Headline result

A small open-weight model with fastWorkflow surpasses frontier baselines on τ-bench at a fraction of the inference cost.

Tags agents · SLMs · tool-use · τ-bench · failure taxonomy

Extracting Insights from Structured Data using Hierarchical Clustering of Itemsets and LLM-Based Summarization

Sanchit Satija, Dhar Rawal
International Conference on Pattern Recognition (ICPR) — 2026

A pattern-discovery framework that pairs frequent itemset mining (FPMax) with hierarchical clustering over a composite structural-and-instance similarity, then uses LLMs for semantic categorization and multi-level narrative generation.

Headline result

Validated on the MIMIC-III clinical database via a blind study with medical experts; LLM categorization ranked in the top two by all three.

Tags pattern mining · FPMax · clustering · MIMIC-III · LLM summarization

  Stack

Languages & Core

Python PyTorch LaTeX

Agents & LLMs

DSPy Transformers LangGraph vLLM

Data & Analysis

pandas scikit-learn SciPy mlxtend


  GitHub Activity


  Reach out

If you're working on agentic systems with small models, principled evaluation of LLM tool use, or LLM-driven knowledge discovery from structured data — I'd love to talk.


Pinned Loading

  1. radiantlogicinc/cluster_rank_summarize radiantlogicinc/cluster_rank_summarize Public

    Clustering, ranking and summarizing structured data

    Python 4 4

  2. radiantlogicinc/fastworkflow radiantlogicinc/fastworkflow Public

    A framework for building large-scale, deterministic, interactive workflows with a fault-tolerant, conversational UX

    Python 48 14

  3. DATABASE-MANAGEMENT-SYSTEM---ZOO-AND-WILDLIFE-DATA DATABASE-MANAGEMENT-SYSTEM---ZOO-AND-WILDLIFE-DATA Public

  4. Exploratory-Data-Analysis-on-Sports Exploratory-Data-Analysis-on-Sports Public

    Jupyter Notebook

  5. fastworkflow fastworkflow Public

    Forked from radiantlogicinc/fastworkflow

    A framework for building large-scale, deterministic, interactive workflows with a fault-tolerant, conversational UX

    Python

  6. cluster_rank_summarize cluster_rank_summarize Public

    Forked from radiantlogicinc/cluster_rank_summarize

    Clustering, ranking and summarizing structured data

    Python 1