Deep Learning Research Engineer. I build frontier AI architectures from scratch in raw PyTorch — no Trainer, no Lightning, every layer by hand — and I write code the way a reviewer would want to read it: documented, tested, and reproducible.
I'm open to Deep Learning, LLM, and GenAI engineering roles (remote or on-site, worldwide).
- Portfolio: https://atandra2000.github.io/mycv
- LinkedIn: https://www.linkedin.com/in/atandrabharati
- Email: atandra.bharati@gmail.com
Self-taught. B.Tech Civil Engineering, 2024. I moved into deep learning by implementing it rather than watching it: starting from GANs and VAEs, up through LLM pretraining, latent diffusion, and multi-agent systems. Civil engineering taught me constraint thinking — models must hold under memory and compute budgets the same way structures hold under load — and that framing shows up in every project below.
| Project | What it is | Details |
|---|---|---|
| HyMo | Hybrid LLM: gated Delta Net (linear attention) × MLA, asymmetric MoE, MTP. Pretrained from scratch on 30B tokens with FSDP-2 and a custom Triton GDN kernel. | 434M active / 1.13B stored params |
| LLaMA-3-Lite | LLaMA 3-style transformer for single-GPU pretraining. GQA, RoPE, Flash Attention, chunked cross-entropy, disk-backed token cache. | 78% peak-memory reduction (92 GB → 20 GB) on a single A100 |
| DeepSeek-v3-Lite | From-scratch reimplementation of DeepSeek-V3: MLA, aux-loss-free MoE, MTP, speculative decoding. Includes a 1,950-line technical reference on MLA. | ~412M params, Chinchilla-optimal 8.4B-token run |
| GPT-OSS-Lite | Reproduction of GPT-OSS: sliding/full attention alternation, learned attention sinks, YaRN 128K, top-2-of-8 MoE. | 502M total / 247M active params |
| Mamba-3-Lite | Mamba-3 reproduction: complex64 SSD, MIMO inter-head mixing, zero causal convolution. Pure PyTorch, no custom CUDA. | ~434M params, 8B-token run |
| TranslationLM | Encoder-decoder transformer for English→Italian translation with word-level tokenizers on opus_books. | Full seq2seq from scratch |
| Project | What it is | Details |
|---|---|---|
| StableDiffusion | Stable Diffusion 1.x-class latent diffusion trained from scratch: full UNet, DDPM/DDIM, DDP + BF16 on 2× RTX 5090. | 860M UNet, 1.3M+ images across 7 phases |
| ActionRecognition | Skeleton-based action recognition: from-scratch HRNet-like pose estimation + two-stream ST-GCN. Served with ONNX, TensorRT, FastAPI. | ~30 FPS on RTX 3090 |
| detect-objects | Real-time detector: ResNet-50+FPN with a deformable set-prediction decoder (RT-DETR/DINO-style), no anchors, no NMS. | COCO 2017, ONNX export |
| upscale-sr | 4× real-world super-resolution: StableSR-style latent-diffusion UNet + mamba-ssm refiner. | ~1.5s per image on RTX 5090 |
| VisionLanguageModel | PaliGemma-inspired VLM: SigLIP encoder, GQA decoder with RoPE, linear projector. Zero pretrained weights. | COCO 2014 |
| FaceAgingCycleGAN | Bidirectional face aging with AdaIN-conditioned CycleGAN, multi-scale discriminator, VGG perceptual loss. | IMDB-Wiki |
| DCGAN-Face-Generation | DCGAN on CelebA with experiment tracking, from scratch. | 64×64 faces |
| FaceGenerationVAE | β-VAE for face generation: bilinear upsampling, KL annealing, 512-dim latent. | CelebA, Comet tracking |
| Project | What it is | Details |
|---|---|---|
| AutonomousResearcher | Multi-agent platform that reads papers, plans experiments, writes and runs training code, and iterates. Provider-agnostic LLM routing, 15 phases, 23 agents, 61 tools. | 878 passing tests |
| news-agent | Autonomous research-intelligence agent: daily multi-source sweep, LLM synthesis with self-critique, provenance-tracked reports. | 291 offline tests |
| Project | What it is |
|---|---|
| LearnAgenticAI | Ten progressive agentic-AI projects with LangChain, LangGraph, LangSmith, MCP, and a Next.js chat UI. |
| FusionLLM | Hybrid LLM pretraining framework: MLA + gated Delta Net + DeepSeek MoE + MTP. 415M active params. |
| GPT-From-Scratch | Character-level GPT on Tiny Shakespeare. The project that started the portfolio. |
| qwen3.8-guide | Interactive first-principles explainer of the Qwen 3.8-27B architecture. |
| flash-attention-2-guide | Interactive FlashAttention-2 field guide: online softmax, SRAM tiling, backward pass. |
- Multi-Head Latent Attention & Mixed Precision — technical deep-dive — 1,950 lines covering KV-cache compression math, the absorption trick, and decoupled RoPE.
- Qwen 3.8-27B architecture guide — derived from first principles.
- FlashAttention-2 guide — online softmax, SRAM tiling, backward pass.
Languages: Python, SQL, JavaScript ML: PyTorch 2.x (BF16/FP16/FP8, SDPA, Flash-Attention 2, torch.compile), HuggingFace, diffusers, CUDA, DDP, FSDP-2 Agentic: LangChain, LangGraph, LangSmith, MCP, pydantic v2 Tracking & serving: W&B, Comet, ONNX, TensorRT, FastAPI Infrastructure: Linux, Docker, Git
- From-scratch PyTorch everywhere; no Trainer, Lightning, or accelerate.
- Every project tracked on W&B or Comet with full RNG-state reproducibility and atomic checkpoint writes.
- Faithful paper reproduction: DeepSeek-V3, LLaMA-3, GPT-OSS, Mamba-3, PaliGemma — implemented to the paper, with regression tests for every bug found.
B.Tech Civil Engineering, Heritage Institute of Technology, Kolkata, 2024. The transition from civil to ML is a strength, not a gap: engineering discipline under constraint — memory budgets, distributed training, reproducibility — transfers directly to deep learning systems.