Ankit Dahal

Projects

  • Browser use agents

    Current VLM · Agents

    Building browser use agents with Claude Opus/Qwen 3.6 depending on task. Repurposed a self-built stealth browser harness originally intended for captcha solving. Currently supports research tasks: answering questions based on multi-document source material. Future plans for coordinated, multi agent web scraping.

  • Lambda Cloud MCP

    MCP · ML infra · Agents

    MCP surface for Lambda Cloud — auto-provision GPUs, orchestrate ML environments and training jobs, steer agents by text message. Code

  • Legal Agentic RAG System

    RAG · Agents

    Agentic RAG on 250k+ legal pages with hybrid search (bge-m3 + BM25), reranking, multi-agent system & orchestration. 10,000+ HuggingFace dataset downloads. Code · Training infra · Writeup

  • LoRA Targeting for Persona SFT

    Fine-tuning · MoE · VLM

    Gradient-guided LoRA targeting for persona SFT on qwen3.5-35b-a3b; lmms-eval OSS contribution (eval harness used by frontier labs & thousands of users). Accelerated evals by 30x with vLLM. Code · Writeup

  • Llama 3.1 8B Instruction Tuning

    Fine-tuning

    LoRA SFT with +52% IFEval improvement (200→305/834) on ~$10 compute, and multiple quantization experiments. Code

  • Large Language Model Pretraining

    Pretraining · DDP

    Pretrained a 450M-param transformer on FineWeb-Edu (10B tokens), 8xA100 gpus using DDP — RoPE, SwiGLU, Flash attention, GQA/MHA, KV-Cache, etc. Code · Checkpoint

  • Transformer from Scratch

    Fundamentals

    PyTorch "Attention Is All You Need" implementation with encoder-only, decoder-only, and encoder-decoder architectures. Along with modern extensions (SwiGLU, RMSNorm, RoPE) Code

Writing

Technical Writeup

LoRA Targeting for Discord-style SFT

Deep-dive on gradient attribution-guided LoRA targeting, dataset curation from raw Discord exports, training/eval loops, and checkpoint selection for authentic persona style transfer without catastrophic forgetting.
Technical Writeup

Agentic RAG with production level accuracy

End-to-end design notes covering hybrid retrieval architecture, embedding strategies, reranker integration, agentic orchestration, and lessons learned building a RAG pipeline on real legal documents with production level accuracy.