Get in touch

Work

Projects at LG Soft India since 2023: from measuring audio–video timing with sensors, to video quality models with IISc, to on-device speech AI, to multi-agent systems for software engineering.

Timeline

Choose a project to see the details.

AgenticMar 2026 – now

Multi-agent framework for GStreamer patch migration

LG Soft India

Port LG's ~4,000 downstream GStreamer patches from 1.24 to 1.28, a job that usually takes about ten person-months.

~46of ~4,000 patches needed a human; the agents handled the rest
Amazon KiroMulti-agentRAG memorygit rerereGStreamer
Read case study →

All projects

  1. Multi-agent framework for GStreamer patch migration

    AgenticMar 2026 – nowLG Soft India

    Port LG's ~4,000 downstream GStreamer patches from 1.24 to 1.28, a job that usually takes about ten person-months.

    • A multi-agent orchestrator on Amazon Kiro that ports the patch series one patch at a time.
    • Per-patch pipelines of specialised agents for conflict detection, API-compatibility analysis, semantic auditing, blast-radius analysis, conflict resolution and test verification.
    • A deep patch analyser that produced a detailed analysis of all ~4,000 patches, stored in a per-patch artifact store keyed by commit SHA and fed to the agents as grounded input during the rebase.
    • Retrieval-augmented (RAG) memory plus git rerere, so conflicts resolved once are reused automatically in future migrations.

    ~46 of ~4,000 patches needed a human; the agents handled the rest

    Amazon KiroMulti-agentRAG memorygit rerereGStreamer
    Read case study →

  2. AI-based live audio content filtering for webOS TVs

    Speech & AudioOct 2025 – Mar 2026LG Soft India

    Mute an offensive word in live TV audio before the viewer hears it, on the TV itself.

    • A real-time pipeline that connects streaming speech analysis to the TV media stack.
    • Changes to GStreamer-based playback pipelines and webOS media services so audio can be analysed and muted in the playback path.
    • An evaluation of lightweight edge AI models under on-device limits, focused on low-latency inference and robustness to broadcast noise.

    Live on-device muting inside the TV's GStreamer playback

    Edge AIStreaming ASRGStreamerwebOS
    Read case study →

  3. Audio–video quality assessment with IISc Bangalore

    Audio & VideoJun 2024 – Dec 2025Academic–industry collaboration

    Predict how people would rate a TV's audio and video quality, so testing doesn't depend on large human studies.

    • Capture setups that route a TV's audio and video output into PC-based analysis pipelines through capture cards, for controlled benchmarking and dataset creation.
    • Multi-stage dataset pipelines that combine real TV output with synthetically generated audio and video degradations.
    • Evaluations of the audio and video quality models proposed by the academic team.

    18 mo research collaboration with IISc

    Quality assessmentCapture pipelinesEmbedding analysisDatasets
    Read case study →

  4. Audio–visual sync and timing measurement for HbbTV

    HardwareFeb 2023 – Jun 2024LG Soft India

    Measure exactly when picture and sound reach the viewer during broadcast ad switches on European TVs.

    • Phototransistor-based optical sensing and analog audio capture with microcontrollers, measuring display timing at millisecond resolution.
    • Measurement protocols for switch-time characterisation in targeted-advertisement scenarios, using QR-based visual markers and audio-decoder events aligned with the HbbTV specifications.
    • A measurement harness scaled with parallel processing and task queues.

    4× measurement throughput, at millisecond resolution

    PhototransistorsMicrocontrollersHbbTVTask queues
    Read case study →