The posting
Research Engineer, Applied AI
Location: Bangalore (or throughout India remote-friendly with travel)
About EnCharge AI:
EnCharge AI is building the next generation AI platform. Our novel in-memory-computing architecture delivers a 10x step-function improvement in compute energy efficiency and performance for AI inference workloads. As the demands of artificial intelligence move beyond today's models, we believe fundamental underlying infrastructure must evolve. We are an experienced team of AI researchers, silicon & systems engineers, and architects backed by leading investors, poised to become the essential platform for the next wave of AI innovation.
The Opportunity:
Modern AI workloads—from large language models to diffusion-based generators to multimodal systems—represent some of the most compute-intensive frontiers in AI, and some of the most promising applications for our hardware’s energy efficiency advantages. We’re building a vertically integrated AI stack that will showcase the transformative potential of our silicon while delivering real value to customers today.
We are seeking a Research Engineer to push the boundaries of AI model capability, quality, and efficiency. You’ll build fine-tuning and post training pipelines, develop rigorous benchmarking frameworks, and work at the intersection of ML research and hardware-aware optimization—ensuring our models run beautifully on our silicon.
This is a role for someone who thrives at the boundary between research and engineering. You’ll read papers, implement techniques, and ship production-quality code—all in service of making AI inference faster, cheaper, and better.
Key Responsibilities:
- Algorithmic Acceleration: Research and implement state-of-the-art techniques to accelerate AI inference—quantization, sparsity, distillation, speculative decoding, caching strategies, and architectural modifications. Systematically characterize tradeoffs between model quality, latency, throughput, and power consumption to find optimal operating points across different use cases.
- Hardware Co-Design: Partner closely with hardware, compiler, and quantization teams to ensure algorithmic improvements translate to real gains on our silicon. Identify optimizations aligned with our architecture's strengths—maximizing throughput while minimizing power. Shape the feedback loop between model development and hardware.
- Evaluation: Build profiling tools and comprehensive benchmarking frameworks to understand compute bottlenecks, measure model quality across standard and domain-specific evals, and track efficiency metrics.
- Applied Research: Build robust fine-tuning workflows for modern AI models, enabling rapid experimentation with LoRA, adapters, and full fine-tuning. Stay current with the rapidly evolving landscape—evaluate new architectures, implement promising techniques, and contribute insights that inform technical and go-to-market strategy.
Qualifications:
- 5+ years of experience in ML research, applied ML, or ML systems
- Strong fundamentals in Python and PyTorch
- Hands-on experience with transformers, diffusion models, state space models etc.
- Experience fine-tuning large models and building training/evaluation pipelines
- Deep understanding of transformers, attention mechanisms, & optimization techniques
- Comfort reading and implementing techniques from research papers
Nice to Have:
- Experience with efficient inference techniques (KV cache optimization, attention variants, MoE routing, flow matching)
- Background in hardware-aware ML optimization or quantization
- Familiarity with profiling tools (PyTorch Profiler, Nsight, custom instrumentation)
- Publications in generative modeling, efficient inference, or ML systems
- Contributions to open-source ML projects
